# MCP Catalogs — Full Index (LLM-friendly) > Comprehensive markdown dump of every active MCP server we index (1989 entries), ranked by composite score. Each row links to a richer detail page with install snippets, FAQ, and side-by-side comparisons. Generated 2026-05-18. Refreshed daily. Composite score = AI evaluation (30%) + log(stars) (30%) + log(real Smithery uses, 30d) (30%) + recency (10%). Official servers receive a 1.15x multiplier. Range: 0–115. Companion files: [https://mcpcatalogs.com/llms.txt](https://mcpcatalogs.com/llms.txt) · [https://mcpcatalogs.com/sitemap.xml](https://mcpcatalogs.com/sitemap.xml) --- ## ai-llm (422) ### [ultimate_mcp_server](https://mcpcatalogs.com/en/server/dicklesworthstone-ultimate-mcp-server) by Dicklesworthstone · score 85 · ★149 Comprehensive MCP server providing dozens of capabilities for AI agents including LLM delegation, browser automation, document processing, and cognitive memory systems. **When to choose:** Choose this MCP server when you need a comprehensive toolkit for AI agents that includes LLM delegation, browser automation, document processing, and cognitive memory systems. **When NOT to choose:** Avoid this server if you need a lightweight, minimal solution or if you require specialized MCP servers that focus on a single capability rather than a comprehensive suite. ### [mcp-server-chart](https://mcpcatalogs.com/en/server/antvis-mcp-server-chart) by antvis · score 84 · ★4,068 · 10,239 uses/30d A TypeScript MCP server for generating 26+ visualization charts using AntV, supporting multiple chart types and deployment options. **When to choose:** Choose this when you need comprehensive chart generation capabilities directly within MCP-enabled applications and are already using or willing to integrate with the AntV ecosystem. **When NOT to choose:** Avoid this if you need geographic visualization outside of China, as the map functionality is limited to Chinese regions using AMap service. ### [sequentialthinking](https://mcpcatalogs.com/en/server/modelcontextprotocol-servers-sequentialthinking) by modelcontextprotocol · score 75 · ★85,748 · official Sequential Thinking MCP Server enables dynamic problem-solving through step-by-step reasoning. **When to choose:** Choose this server when working on complex problems requiring structured reasoning, planning tasks that may evolve, or debugging issues that benefit from step-by-step analysis. **When NOT to choose:** Don't choose this for simple problems that can be solved in one step or when you need immediate, non-iterative responses without maintaining reasoning context. ### [context-mode](https://mcpcatalogs.com/en/server/mksglu-context-mode) by mksglu · score 62 · ★14,865 MCP server that optimizes context windows for AI agents by sandboxing tool output with 98% reduction. **When to choose:** Choose Context Mode when working with large codebases or long coding sessions where context overflow is a bottleneck, and especially when using supported platforms like Claude Code or Gemini CLI. **When NOT to choose:** Avoid if you need full raw tool outputs in your context, or if you're using an unsupported platform that requires manual implementation of all routing logic. ### [gpt-researcher](https://mcpcatalogs.com/en/server/assafelovic-gpt-researcher) by assafelovic · score 61 · ★27,100 An autonomous research agent that conducts deep research using any LLM providers with MCP server integration for specialized data sources. **When to choose:** Choose this when you need a comprehensive research tool that can integrate with multiple data sources through MCP, especially for generating detailed reports with citations. **When NOT to choose:** Don't choose this if you need an actual MCP server that exposes tools/resources, rather than a client that consumes them, or if you need write access capabilities. ### [Skill_Seekers](https://mcpcatalogs.com/en/server/yusufkaraaslan-skill-seekers) by yusufkaraaslan · score 61 · ★13,581 A comprehensive MCP server that converts documentation, GitHub repos, and PDFs into Claude AI skills with automatic conflict detection. **When to choose:** Choose this when you need to preprocess multiple types of documentation and codebases into structured knowledge assets for AI systems like Claude or RAG pipelines, especially when you need comprehensive skills with automatic examples and metadata. **When NOT to choose:** Don't choose this if you need write access to your data sources, as this is a read-only preprocessing tool, or if you're already invested in a different documentation-to-AI pipeline with established workflows. ### [OpenMetadata](https://mcpcatalogs.com/en/server/open-metadata-openmetadata) by open-metadata · score 59 · ★13,935 OpenMetadata provides an MCP server for AI assistants to access unified metadata context for data discovery, governance, and lineage. **When to choose:** Choose OpenMetadata MCP when you need comprehensive data governance and context for AI assistants, especially with complex data landscapes requiring lineage and semantic understanding. **When NOT to choose:** Avoid if you need a lightweight solution for simple metadata access or if you're not prepared to deploy and maintain a full metadata platform. ### [mcp-agent](https://mcpcatalogs.com/en/server/lastmile-ai-mcp-agent) by lastmile-ai · score 59 · ★8,325 A comprehensive Python framework for building effective agents with Model Context Protocol and simple workflow patterns. **When to choose:** Choose mcp-agent when you need to build production-ready LLM agents with advanced patterns like durable execution, human-in-the-loop, and cloud deployment. **When NOT to choose:** Avoid if you need a minimal MCP implementation or are already deeply invested in another agent framework with different abstractions. ### [osaurus](https://mcpcatalogs.com/en/server/osaurus-ai-osaurus) by osaurus-ai · score 59 · ★5,287 A native macOS AI harness with MCP server support for local/cloud models, agents, and tools. **When to choose:** Choose Osaurus if you want a native macOS AI agent with persistent memory, sandboxed execution, and MCP server capabilities that works both offline and with cloud providers. **When NOT to choose:** Don't choose Osaurus if you need cross-platform support or are looking for an open-source alternative with more flexible licensing options. ### [bifrost](https://mcpcatalogs.com/en/server/maximhq-bifrost) by maximhq · score 58 · ★4,961 Bifrost is a high-performance AI gateway with MCP support for tool calling, load balancing, and semantic caching across 1000+ models. **When to choose:** Choose Bifrost when you need a high-performance unified gateway for multiple AI providers and want MCP capabilities as part of a comprehensive AI infrastructure solution. **When NOT to choose:** Don't choose Bifrost if you specifically need a dedicated MCP server for direct tool integration, or if you prefer open-source solutions with no vendor dependencies. ### [openagent](https://mcpcatalogs.com/en/server/the-open-agent-openagent) by the-open-agent · score 58 · ★4,760 OpenAgent is a comprehensive AI assistant platform supporting MCP tool integration and agent automation. **When to choose:** Choose OpenAgent when you need a comprehensive personal AI assistant platform with workflow automation, document processing, and support for multiple LLM providers, including MCP integration as one feature. **When NOT to choose:** Don't choose OpenAgent if you're specifically looking for a dedicated MCP server; this is a full platform with many features beyond MCP integration. ### [arxiv-mcp-server](https://mcpcatalogs.com/en/server/blazickjp-arxiv-mcp-server) by blazickjp · score 58 · ★2,738 An MCP server for searching and analyzing ArXiv papers with local storage and advanced search capabilities. **When to choose:** Choose this server for research-focused AI applications that need access to academic papers from arXiv, with proper search capabilities and local caching. **When NOT to choose:** Don't choose this if you need write access to papers, as it only provides read-only access, or if you require processing papers from other academic repositories. ### [mcp-memory-service](https://mcpcatalogs.com/en/server/doobidoo-mcp-memory-service) by doobidoo · score 56 · ★1,848 Persistent memory service for AI agents with REST API, MCP, OAuth, and knowledge graph functionality. **When to choose:** Choose this when you need a self-hosted memory solution for multiple AI agents with shared knowledge graphs and browser-based Claude AI support. **When NOT to choose:** Not ideal if you need a simple, lightweight solution or if you prefer cloud-hosted services over self-hosted infrastructure. ### [mcp-brasil](https://mcpcatalogs.com/en/server/mcp-brasil-mcp-brasil) by Mcp-Brasil · score 56 · ★1,561 Comprehensive MCP server providing access to 70 Brazilian public APIs with 533 tools for various government data sources. **When to choose:** Choose this MCP server when you need comprehensive access to Brazilian public data for AI applications, especially if you're working with economic, legislative, or transparency data. **When NOT to choose:** Don't choose this if you need access to non-Brazilian data sources or if you require write capabilities to the underlying APIs, as this server only provides read access. ### [py-xiaozhi](https://mcpcatalogs.com/en/server/huangjunsen0406-py-xiaozhi) by huangjunsen0406 · score 56 · ★3,309 A Python-based Xiaozhi AI voice assistant with MCP tools integration, supporting GUI/CLI/GPI interfaces and multimodal interactions. **When to choose:** Choose py-xiaozhi for building voice-controlled applications with multimodal AI capabilities, especially when you need cross-platform support without hardware dependencies. **When NOT to choose:** Avoid if you need a pure MCP server without voice interaction capabilities, or if you prefer cloud-based AI services rather than local deployment. ### [Dive](https://mcpcatalogs.com/en/server/openagentplatform-dive) by OpenAgentPlatform · score 56 · ★1,792 Dive is a cross-platform desktop application serving as an MCP host for integrating various LLMs with AI agent capabilities. **When to choose:** Choose Dive when you need a desktop interface to manage MCP servers across multiple LLM providers without complex local deployments. **When NOT to choose:** Don't choose Dive if you need MCP server authentication (currently unstable) or require lightweight resource usage with minimal dependencies. ### [MiniMax-MCP](https://mcpcatalogs.com/en/server/minimax-ai-minimax-mcp) by MiniMax-AI · score 56 · ★1,476 Official MCP server for MiniMax's Text-to-Speech, image and video generation APIs. **When to choose:** Choose this when you need multimedia generation capabilities (speech, images, videos) integrated into your MCP client workflow and you're already using MiniMax's API. **When NOT to choose:** Avoid if you need open-source alternatives, want to avoid vendor lock-in with MiniMax, or need more extensive multimedia editing capabilities beyond generation. ### [korean-law-mcp](https://mcpcatalogs.com/en/server/chrisryugj-korean-law-mcp) by chrisryugj · score 56 · ★1,732 Korean legal MCP server transforming 41 government APIs into 17 tools with citation verification, impact mapping, and time comparison. **When to choose:** Choose this server if you need Korean legal information and analysis in AI applications, particularly when working with Korean law, verifying legal citations, or understanding legal impacts across time periods. **When NOT to choose:** Don't choose this if you need legal information for countries other than South Korea, require write access to legal databases, or need authentication beyond the API key model. ### [paper-search-mcp](https://mcpcatalogs.com/en/server/openags-paper-search-mcp) by openags · score 56 · ★1,461 MCP server for searching and downloading academic papers from multiple sources with free-first strategy. **When to choose:** Choose Paper Search MCP when you need comprehensive academic paper search capabilities across multiple repositories with a free-first approach and standardized output formats for AI integration. **When NOT to choose:** Don't choose this if you need access to paid-only journals (IEEE Xplore, ACM DL require API keys and are still in skeleton stage) or if your institution restricts access to many of the supported open repositories. ### [deep-research](https://mcpcatalogs.com/en/server/u14app-deep-research) by u14app · score 56 · ★4,583 An MCP server for deep research using multiple LLMs with web search capabilities and local processing. **When to choose:** Choose this tool when you need rapid, in-depth research with multiple LLM providers and want to integrate deep research capabilities via MCP into other AI workflows. **When NOT to choose:** Don't choose this if you need real-time data analysis capabilities or if you require write access to external systems as it's primarily focused on research report generation. ### [PrismerCloud](https://mcpcatalogs.com/en/server/prismer-ai-prismercloud) by Prismer-AI · score 55 · ★1,529 Prismer Cloud MCP server provides 47 tools for agent evolution, memory management, community features, and skill marketplace integration. **When to choose:** Choose Prismer Cloud when building AI agents that need to learn from experience, share knowledge across multiple agents, and require robust memory and context management. **When NOT to choose:** Avoid if you need open-source or self-hosted solution, as Prismer Cloud is a proprietary service requiring API keys and external dependencies. ### [mcp-gsc](https://mcpcatalogs.com/en/server/aminforou-mcp-gsc) by AminForou · score 55 · ★843 MCP server connecting Google Search Console to AI assistants for SEO analysis through natural language conversations. **When to choose:** Choose this if you're an SEO professional who wants to analyze Google Search Console data through conversational AI without switching between dashboards and reports. **When NOT to choose:** Don't choose this if you need real-time data (GSC has a 2-3 day lag) or if you need write access to your Search Console properties (this server is read-only). ### [MeiGen-AI-Design-MCP](https://mcpcatalogs.com/en/server/jau123-meigen-ai-design-mcp) by jau123 · score 55 · ★1,108 MCP server for AI image & video generation with 9 models, 1,400+ prompts, and parallel sub-agent orchestration. **When to choose:** Choose MeiGen when you need professional-grade AI image and video generation with multiple models, curated prompts, and parallel batch processing directly within your AI coding environment. **When NOT to choose:** Avoid MeiGen if you need maximum control over generation parameters without abstraction layers, or if you prefer to work directly with image generation APIs rather than through an MCP server intermediary. ### [datagouv-mcp](https://mcpcatalogs.com/en/server/datagouv-datagouv-mcp) by datagouv · score 55 · ★1,460 Official MCP server for data.gouv.fr that enables AI chatbots to search and analyze French open data. **When to choose:** Choose this server when you need conversational access to French public datasets from data.gouv.fr, especially if you're working with Claude, ChatGPT, or other major AI chat platforms. **When NOT to choose:** Don't choose this if you need access to non-French datasets or require write capabilities to modify datasets, as this server is read-only focused on French government data. ### [nocturne_memory](https://mcpcatalogs.com/en/server/dataojitori-nocturne-memory) by Dataojitori · score 55 · ★1,080 A long-term memory server for MCP agents that enables persistent, cross-session identity and context. **When to choose:** Choose Nocturne Memory when you want your AI agents to maintain persistent, self-structured memories across different models and sessions, with visual management capabilities for humans. **When NOT to choose:** Don't choose this if you need simple, short-term context memory or if you're looking for a fully managed SaaS solution without self-hosting requirements. ### [zenfeed](https://mcpcatalogs.com/en/server/glidea-zenfeed) by glidea · score 54 · ★1,677 An AI-powered RSS reader that can function as an MCP server for intelligent feed management. **When to choose:** Choose zenfeed as your MCP server when you need an AI-powered information processing system that can monitor multiple RSS feeds, filter content based on custom rules, and deliver personalized briefings or notifications. **When NOT to choose:** Avoid zenfeed if you need a system with robust authentication mechanisms, as it currently lacks security features and isn't recommended for public deployment without proper firewall configuration. ### [paperbanana](https://mcpcatalogs.com/en/server/llmsresearch-paperbanana) by llmsresearch · score 54 · ★1,549 PaperBanana is an MCP server that generates academic diagrams and research visuals from text descriptions. **When to choose:** Choose PaperBanana if you need automated generation of academic diagrams and statistical plots from text descriptions, especially for research papers and presentations. **When NOT to choose:** Don't choose PaperBanana if you need real-time collaboration features or if you're working on a project with extremely tight visual branding requirements that can't be adjusted. ### [zotero-mcp](https://mcpcatalogs.com/en/server/cookjohn-zotero-mcp) by cookjohn · score 54 · ★784 Zotero MCP plugin integrates AI assistants with Zotero reference management via MCP protocol, providing search, annotation analysis, content extraction, and write operations. **When to choose:** Choose Zotero MCP if you work with academic literature and want AI assistants to directly interact with your Zotero library for advanced search, content analysis, and note management. **When NOT to choose:** Don't choose Zotero MCP if you need read/write access to cloud-only references without local PDF storage, or if you're using Zotero 6 or earlier versions. ### [gemini-mcp-tool](https://mcpcatalogs.com/en/server/jamubc-gemini-mcp-tool) by jamubc · score 54 · ★2,212 MCP server enabling AI assistants to interact with Google Gemini CLI for large file analysis and codebase understanding. **When to choose:** Choose this MCP server when you need to analyze large codebases or files using Gemini's capabilities directly in Claude, especially when working with projects that exceed standard context windows. **When NOT to choose:** Avoid this if you need write access to your codebase (Gemini CLI is read-only) or if you're already using other Google AI services that provide similar functionality. ### [reddit-mcp-buddy](https://mcpcatalogs.com/en/server/karanb192-reddit-mcp-buddy) by karanb192 · score 54 · ★649 A production-ready MCP server for browsing Reddit with AI assistants, featuring multiple auth tiers and clean data. **When to choose:** Choose this server when you need clean Reddit data access for AI assistants without API registration hassles, especially for trend analysis and sentiment tracking across subreddits. **When NOT to choose:** Don't choose this if you need write access to Reddit (this is read-only only) or require access to private subreddits with restricted access. ### [MCP](https://mcpcatalogs.com/en/server/jina-ai-mcp) by jina-ai · score 54 · ★680 Jina AI Remote MCP server provides web search, content reading, and AI tools with URL filtering capabilities. **When to choose:** Choose this server when you need comprehensive web scraping, search, and content processing capabilities, especially when working with academic research or requiring both single and parallel processing of web content. **When NOT to choose:** Avoid this server if you need offline functionality, have concerns about vendor lock-in with Jina AI, or require extensive customization beyond what the API provides. ### [skybridge](https://mcpcatalogs.com/en/server/alpic-ai-skybridge) by alpic-ai · score 53 · ★1,128 Skybridge is a TypeScript framework for building MCP and ChatGPT apps with type safety and React components. **When to choose:** Choose Skybridge when building complex AI applications with React interfaces that need to work with both ChatGPT and MCP clients. **When NOT to choose:** Don't choose Skybridge if you need a simple MCP server implementation or if you're not building a React-based AI application. ### [mirobody](https://mcpcatalogs.com/en/server/thetahealth-mirobody) by thetahealth · score 53 · ★1,042 A personal data platform bridging health, finance data with AI via MCP servers and Claude Agent Skills. **When to choose:** Choose Mirobody for health-focused personal AI applications that need to integrate with multiple health devices and standards while maintaining user privacy. **When NOT to choose:** Don't choose Mirobody if you need general-purpose AI without health data integration, or if you prefer a simpler deployment without Docker or PostgreSQL. ### [better-chatbot](https://mcpcatalogs.com/en/server/cgoinglove-better-chatbot) by cgoinglove · score 53 · ★1,099 An open-source AI chatbot with MCP tool integration, voice assistant, and workflow capabilities. **When to choose:** Choose this if you need a chatbot interface that integrates multiple AI providers and MCP tools for enhanced functionality. **When NOT to choose:** Don't choose this if you're looking for a standalone MCP server or a tool that doesn't require a full chatbot application. ### [MassGen](https://mcpcatalogs.com/en/server/massgen-massgen) by massgen · score 53 · ★1,015 Multi-agent scaling system using parallel refinement and voting for higher quality AI results. **When to choose:** Choose MassGen when you need collaborative problem-solving with multiple AI models, especially when high-quality output through consensus and refinement is critical. **When NOT to choose:** Don't choose MassGen for simple, single-model tasks or if you need minimal setup complexity as the system requires multiple API keys and configuration. ### [wren-engine](https://mcpcatalogs.com/en/server/canner-wren-engine) by Canner · score 53 · ★661 An open context engine for AI agents that provides business context and semantic layer over data sources. **When to choose:** Choose Wren Engine when your AI agents need to understand business context and metrics beyond raw table schemas, especially for governed enterprise data environments. **When NOT to choose:** Don't choose this if you need direct database access without business context layer or if you're looking for a simple text-to-SQL solution without semantic modeling. ### [Wax](https://mcpcatalogs.com/en/server/christopherkarani-wax) by christopherkarani · score 53 · ★734 Swift-native memory layer for AI agents providing high-performance RAG on Apple Silicon with zero configuration. **When to choose:** Choose Wax when building Apple-specific AI agents that need fast, private memory without cloud dependencies or complex setup. **When NOT to choose:** Don't choose Wax if you need cross-platform support, cloud-based deployment, or have non-Apple hardware requirements. ### [memora](https://mcpcatalogs.com/en/server/agentic-box-memora) by agentic-box · score 53 · ★405 A lightweight MCP server for semantic memory storage, knowledge graphs, and persistent memory for AI agents. **When to choose:** Choose Memora when you need persistent memory for AI agents with semantic search, knowledge graphs, and cross-session context, especially if you're already using cloud services like Cloudflare D1 or S3. **When NOT to choose:** Avoid Memora if you need a self-contained solution without external dependencies, as it requires cloud services like D1 or S3 for optimal functionality, and may not be ideal for highly sensitive data due to cloud storage requirements. ### [vestige](https://mcpcatalogs.com/en/server/samvallad33-vestige) by samvallad33 · score 53 · ★531 Cognitive memory for AI agents with FSRS-6 spaced repetition, 29 brain modules, and a 3D visualization dashboard. **When to choose:** Choose Vestige when you need persistent memory for AI agents with advanced cognitive features like spaced repetition, memory suppression, and neural visualization. **When NOT to choose:** Don't choose Vestige if you need cloud-based memory solutions or if you're working with AI frameworks that don't support MCP integration. ### [haiku.rag](https://mcpcatalogs.com/en/server/ggozad-haiku-rag) by ggozad · score 53 · ★524 Agentic RAG with multimodal search, analysis agents, and MCP server capabilities. **When to choose:** Choose Haiku RAG when you need a production-ready multimodal RAG system with agent capabilities and MCP integration for AI assistants. **When NOT to choose:** Don't choose this if you need a simple vector search without agent capabilities or if you're looking for an open-source solution without external dependencies. ### [Ori-Mnemos](https://mcpcatalogs.com/en/server/aayoawoyemi-ori-mnemos) by aayoawoyemi · score 52 · ★301 Local-first persistent memory system for AI agents with recursive memory harness and MCP integration. **When to choose:** Choose Ori when you need persistent memory for AI agents with sophisticated retrieval capabilities, especially for complex reasoning tasks and long-term conversational memory where you want to maintain complete control over your data. **When NOT to choose:** Don't choose Ori if you need cloud-based memory services with built-in API access, or if you're looking for a simple note-taking solution without the complex graph structure and learning mechanisms. ### [agency-orchestrator](https://mcpcatalogs.com/en/server/jnmetacode-agency-orchestrator) by jnMetaCode · score 52 · ★859 A multi-AI-role orchestration tool that enables complex task completion through collaboration of 211+ expert roles with zero-code YAML configuration. **When to choose:** Choose Agency Orchestrator when you need to orchestrate multiple specialized AI agents for complex workflows without programming, especially when you already have subscriptions to major LLM providers like Claude, ChatGPT Plus, or GitHub Copilot. **When NOT to choose:** Don't choose this if you need low-latency responses or real-time interaction, as it's designed for batch processing of complex multi-agent workflows rather than conversational AI applications. ### [mcp-google-map](https://mcpcatalogs.com/en/server/cablate-mcp-google-map) by cablate · score 52 · ★302 A comprehensive MCP server with 18 Google Maps API tools for geolocation, routing, search, and location-based intelligence. **When to choose:** Choose this MCP server when you need comprehensive Google Maps functionality for AI agents, especially for travel planning, local search, or geographic analysis with 18 specialized tools. **When NOT to choose:** Don't choose if you need a lightweight solution with fewer tools, or if you prefer Google's official Grounding Lite MCP server which offers managed service but with limited functionality. ### [wikipedia-mcp](https://mcpcatalogs.com/en/server/rudra-ravi-wikipedia-mcp) by Rudra-ravi · score 52 · ★250 An MCP server providing Wikipedia access with search, article retrieval, summaries, and multi-language support. **When to choose:** Choose this server when your AI application needs real-time, factual information from Wikipedia and you require multi-language support with various transport protocols. **When NOT to choose:** Don't choose this if you need offline access or require Wikipedia content with specific customizations not available through the MCP API, as it depends entirely on the live Wikipedia API. ### [minima](https://mcpcatalogs.com/en/server/dmayboroda-minima) by dmayboroda · score 52 · ★1,048 Minima is an on-premises RAG system with MCP integration for secure local document querying. **When to choose:** Choose Minima when you need secure, on-premises document retrieval with RAG capabilities and want to integrate with AI assistants like Claude or ChatGPT. **When NOT to choose:** Don't choose Minima if you need advanced document formatting preservation, require real-time collaboration features, or need a fully managed cloud service without self-hosting requirements. ### [AgentChat](https://mcpcatalogs.com/en/server/shy2593666979-agentchat) by Shy2593666979 · score 52 · ★715 AgentChat is an LLM-based intelligent agent communication platform with MCP protocol support for dynamic tool calling and knowledge retrieval. **When to choose:** Choose AgentChat when you need an enterprise-ready platform for building multi-agent systems with human oversight in MCP server generation, especially if you're already using FastAPI and need integration with vector databases like Milvus or ChromaDB. **When NOT to choose:** Avoid AgentChat if you need a simple, lightweight MCP solution without the overhead of a full-stack application, or if you require database backends other than MySQL/Redis with vector stores. ### [gemini-skill](https://mcpcatalogs.com/en/server/wjz-p-gemini-skill) by WJZ-P · score 52 · ★813 MCP server that automates Google Gemini through browser automation for image generation, chat, and image extraction. **When to choose:** Choose this MCP server when you need AI-powered image generation through Gemini within your AI agent workflow and require persistent browser automation with stealth capabilities. **When NOT to choose:** Avoid this if you need direct API access to Gemini without browser automation, or if you require multi-browser parallel processing which is not yet supported. ### [llmwiki](https://mcpcatalogs.com/en/server/lucasastorian-llmwiki) by lucasastorian · score 52 · ★908 LLM Wiki is an MCP server that indexes local documents and enables Claude to build and maintain a knowledge wiki. **When to choose:** Choose LLM Wiki when you need a local, self-hosted solution for creating and maintaining dynamic knowledge bases from your research materials, especially if you want Claude to handle the tedious maintenance of cross-references and citations. **When NOT to choose:** Avoid this if you need vector search capabilities (only basic full-text search in local mode), have multiple users working simultaneously on the same workspace, or prefer a cloud-based solution without self-hosting requirements. ### [Gearsystem](https://mcpcatalogs.com/en/server/drhelius-gearsystem) by drhelius · score 52 · ★367 Accurate Sega Master System emulator with embedded MCP server for AI-assisted debugging. **When to choose:** Choose this when you need AI-assisted debugging for retro game development or reverse engineering on Sega Master System/Game Gear platforms. **When NOT to choose:** Don't choose this for modern game development or if you need support for systems beyond Sega's 8-bit consoles. ### [mineru-tianshu](https://mcpcatalogs.com/en/server/magicyuan876-mineru-tianshu) by magicyuan876 · score 51 · ★634 Enterprise-grade AI preprocessing platform with MCP protocol for document, image, audio, and video processing. **When to choose:** Choose this platform for enterprise document processing pipelines requiring multimodal data preprocessing and MCP integration, especially when working with scientific or professional documents needing high-fidelity conversion. **When NOT to choose:** Avoid if you need lightweight document processing without GPU acceleration or prefer solutions with simpler deployment models. The platform may be overkill for basic document conversion needs. ### [shodh-memory](https://mcpcatalogs.com/en/server/varun29ankus-shodh-memory) by varun29ankuS · score 51 · ★212 Persistent cognitive memory for AI agents and robots with offline support and 37 MCP tools. **When to choose:** Choose Shodh-Memory when you need a persistent memory system for AI agents that works fully offline with minimal dependencies, particularly for robotics applications or when low-latency memory operations are critical. **When NOT to choose:** Avoid this if you need cloud-based memory sync across devices, require integration with OpenAI services, or need a system that already has established enterprise support and documentation. ### [CoexistAI](https://mcpcatalogs.com/en/server/spthole-coexistai) by SPThole · score 51 · ★481 Modular research assistant framework with MCP server integration for web search, YouTube, Reddit, mapping, and GitHub exploration. **When to choose:** Choose CoexistAI when you need a comprehensive research assistant that can gather information from multiple sources including web, social media, and code repositories with a unified interface. **When NOT to choose:** Avoid if you need real-time data processing with low latency or if you're looking for a specialized tool focused on a single information source rather than a multi-platform framework. ### [YourMemory](https://mcpcatalogs.com/en/server/sachitrafa-yourmemory) by sachitrafa · score 51 · ★220 Agentic AI memory system with Ebbinghaus forgetting curve decay, improving recall over competitors. **When to choose:** Choose YourMemory when you need persistent memory for AI agents with biologically-inspired decay and superior recall capabilities compared to alternatives like Mem0 or Zep. **When NOT to choose:** Not suitable for large-scale deployments due to its local-first architecture; also limited by CC BY-NC 4.0 license which prohibits commercial use. ### [search-console-mcp](https://mcpcatalogs.com/en/server/saurabhsharma2u-search-console-mcp) by saurabhsharma2u · score 51 · ★129 MCP server unifying Google Search Console, Bing Webmaster Tools, and GA4 data for AI-driven SEO analytics. **When to choose:** Choose this MCP server when you need AI agents to perform comprehensive SEO analysis across multiple search platforms without manual data exports. **When NOT to choose:** Don't choose this if you need search platform access beyond Google, Bing, and GA4, or if you require real-time data processing capabilities (the server appears to fetch data on-demand). ### [AgentRecall-MCP](https://mcpcatalogs.com/en/server/goldentrii-agentrecall-mcp) by Goldentrii · score 51 · ★252 AgentRecall provides AI session memory with Think-Execute-Reflect quality loops to enable persistent, compounding memory for AI agents. **When to choose:** Choose AgentRecall when you need persistent memory for AI agents across sessions, especially if you're already working with TypeScript-based development projects and want automatic learning from corrections. **When NOT to choose:** Avoid if you need strict isolation between sessions or prefer cloud-only solutions without local data storage. ### [minecraft-mcp-server](https://mcpcatalogs.com/en/server/yuniko-software-minecraft-mcp-server) by yuniko-software · score 51 · ★578 Minecraft MCP server enabling AI assistants to control in-game characters through Mineflayer API. **When to choose:** Choose this MCP server if you're looking to integrate AI capabilities with Minecraft for educational purposes, demonstrations of AI in game environments, or autonomous exploration and building within Minecraft. **When NOT to choose:** Don't choose this server if you need support for Minecraft versions beyond 1.21.11, or if you're looking for production-ready automation with robust error handling and extensive feature set. ### [swarmclaw](https://mcpcatalogs.com/en/server/swarmclawai-swarmclaw) by swarmclawai · score 51 · ★486 SwarmClaw is a self-hosted multi-agent framework with MCP tools for autonomous agent swarms. **When to choose:** Choose SwarmClaw when you need a self-hosted multi-agent framework with memory, delegation, and MCP tools integration across multiple LLM providers. **When NOT to choose:** Avoid SwarmClaw if you need a simple MCP server without the overhead of a full multi-agent framework, or if you prefer cloud-hosted solutions. ### [pasal](https://mcpcatalogs.com/en/server/ilhamfp-pasal) by ilhamfp · score 51 · ★236 MCP server providing grounded access to Indonesian legal data with search, article retrieval, and status checking tools. **When to choose:** Choose Pasal.id when you need accurate, up-to-date information about Indonesian law with proper citations and want to avoid AI hallucinations in legal contexts. **When NOT to choose:** Don't choose Pasal.id if you need legal information for countries other than Indonesia, as it exclusively covers Indonesian laws and regulations. ### [n8n-claw](https://mcpcatalogs.com/en/server/freddy-schuetz-n8n-claw) by freddy-schuetz · score 51 · ★436 Self-hosted AI agent using n8n with MCP Skills library, memory management, and expert agents. **When to choose:** Choose this if you want a complete AI agent solution with task management and memory capabilities that integrates MCP skills as one feature. **When NOT to choose:** Don't choose this if you specifically need an MCP server to expose tools/resources to other systems; this is a consumer agent, not a provider of MCP services. ### [MARM-Systems](https://mcpcatalogs.com/en/server/lyellr88-marm-systems) by Lyellr88 · score 51 · ★288 MARM MCP server provides persistent memory and structured session context for AI agents across tools. **When to choose:** Choose MARM when you need persistent memory across multiple AI sessions and different AI clients, or when working on complex projects requiring continuity of context and multi-agent coordination. **When NOT to choose:** Avoid MARM if you need simple, short-term memory solutions or if you're concerned about the development-only authentication model which requires production OAuth setup. ### [gemini-mcp](https://mcpcatalogs.com/en/server/rlabs-inc-gemini-mcp) by RLabs-Inc · score 51 · ★196 MCP Server enabling Claude to interact with Google's Gemini models, offering text processing, image generation, video creation, and document analysis. **When to choose:** Choose gemini-mcp when you need to integrate Gemini's capabilities (especially image/video generation) into Claude workflows, or when you want a versatile AI assistant with both text and media generation tools. **When NOT to choose:** Avoid this if you need commercial support, as it's a community project without formal backing, or if you're concerned about vendor lock-in to Google's AI ecosystem. ### [boss-agent-cli](https://mcpcatalogs.com/en/server/can4hou6joeng4-boss-agent-cli) by can4hou6joeng4 · score 51 · ★568 An AI-agent CLI tool for BOSS Zhipin job platform with MCP integration. **When to choose:** Choose this if you need an AI agent to automate job searching and application on Chinese recruitment platforms like BOSS Zhipin, with structured data outputs and multi-platform support. **When NOT to choose:** Not ideal for non-Chinese recruitment platforms, or if you need advanced CRM features beyond basic recruitment workflows. ### [vibe-check-mcp-server](https://mcpcatalogs.com/en/server/pv-bhat-vibe-check-mcp-server) by PV-Bhat · score 51 · ★487 Vibe Check MCP provides mentor-like feedback to AI Agents, preventing tunnel-vision and over-engineering through chain-pattern interrupts. **When to choose:** Choose this when your agents tend to over-engineer solutions or get stuck in reasoning loops, especially for complex coding tasks or ambiguous workflows. **When NOT to choose:** Don't use if you need a long-term actively maintained project, as this is in maintenance mode with no further updates planned. ### [ClaudeR](https://mcpcatalogs.com/en/server/imnmv-clauder) by IMNMV · score 51 · ★227 ClaudeR bridges RStudio with MCP-based LLM agents for interactive coding, multi-agent orchestration, and automated manuscript auditing. **When to choose:** Choose ClaudeR when you need to connect RStudio to multiple LLM agents for interactive coding, manuscript auditing, or multi-agent data analysis workflows. **When NOT to choose:** Don't choose ClaudeR if you need write access to external systems (it's primarily focused on R execution and analysis), or if you're not working in the R/RStudio ecosystem. ### [MCP-Bridge](https://mcpcatalogs.com/en/server/secretiveshell-mcp-bridge) by SecretiveShell · score 51 · ★925 MCP-Bridge acts as middleware to provide OpenAI-compatible endpoints that call MCP tools. **When to choose:** Choose MCP-Bridge when you have clients that only support OpenAI API but need to access MCP tools, especially if you're already using Open WebUI v0.6.31+. **When NOT to choose:** Don't choose MCP-Bridge if you're using Open WebUI v0.6.31+ as it now has native MCP support, making MCP-Bridge soft deprecated. ### [solon-ai](https://mcpcatalogs.com/en/server/opensolon-solon-ai) by opensolon · score 50 · ★385 Solon-AI is a Java framework for building AI applications with MCP integration, supporting various LLM providers and multi-agent systems. **When to choose:** Choose Solon-AI when you need a Java-based framework for AI development that integrates MCP protocol support and can work with existing Java infrastructure. **When NOT to choose:** Avoid if you're working primarily in non-Java ecosystems or need a lightweight solution with minimal dependencies. ### [superlocalmemory](https://mcpcatalogs.com/en/server/qualixar-superlocalmemory) by qualixar · score 50 · ★150 A local-only AI memory system with MCP integration for unlimited memory across AI clients. **When to choose:** Choose SuperLocalMemory when you need persistent memory for AI agents with strict privacy requirements and cannot use cloud services. Ideal for developers working with sensitive code or requiring offline functionality. **When NOT to choose:** Not recommended if you need cloud-based memory sharing across multiple devices, or if you prefer cloud-hosted solutions with simpler setup. The AGPL-3.0 license may also be restrictive for some commercial use cases. ### [mcp-server-12306](https://mcpcatalogs.com/en/server/drfccv-mcp-server-12306) by drfccv · score 50 · ★327 A high-performance 12306 train ticket query system that provides real-time data via MCP protocol. **When to choose:** Choose this MCP server for AI applications needing real-time Chinese train ticket information, especially when integrating with Claude Desktop or requiring low-latency access to 12306 data. **When NOT to choose:** Avoid if you need commercial use, international train data beyond China, or require write capabilities to modify train schedules or bookings. ### [Octopoda-OS](https://mcpcatalogs.com/en/server/ryjoxtechnologies-octopoda-os) by RyjoxTechnologies · score 50 · ★323 Octopoda is an MCP server that provides persistent memory, loop detection, and audit trails for AI agents with 29 tools. **When to choose:** Choose Octopoda when you need persistent memory for AI agents, loop detection to prevent wasted tokens, and comprehensive observability—especially if you're already working with Python-based agent frameworks. **When NOT to choose:** Don't choose Octopoda if you're not using Python-based AI frameworks or if you prefer a fully cloud-based solution with no local storage option. ### [daymon](https://mcpcatalogs.com/en/server/daymonio-daymon) by daymonio · score 50 · ★373 Daymon is an MCP server that extends Claude with persistent memory, scheduled tasks, and file watching capabilities. **When to choose:** Choose Daymon if you're already using Claude Desktop or Claude Code and want to extend it with persistent memory, scheduled tasks, and file watching without API costs or cloud dependencies. **When NOT to choose:** Don't choose Daymon if you need to work with AI assistants other than Claude, or if you require cloud-based task automation with multi-device synchronization. ### [nerve](https://mcpcatalogs.com/en/server/evilsocket-nerve) by evilsocket · score 50 · ★1,320 Nerve is an Agent Development Kit with native MCP support, allowing agents to be defined in YAML and act as both MCP client and server. **When to choose:** Choose Nerve if you're looking for a simple, declarative YAML-based approach to building teams of LLM agents with native MCP support for deep orchestration. **When NOT to choose:** Avoid Nerve if you need a solution with extensive GUI interfaces or if the GPL 3 license conflicts with your project's requirements. ### [nexus](https://mcpcatalogs.com/en/server/nexus-router-nexus) by Nexus-Router · score 50 · ★427 Nexus is a unified gateway that aggregates MCP servers, APIs, and LLM providers with security, governance, and routing capabilities. **When to choose:** Choose Nexus when you need a secure gateway to consolidate multiple MCP servers and LLM providers with advanced governance features. **When NOT to choose:** Don't choose Nexus if you need a lightweight solution for connecting to a single MCP server or prefer self-hosted open-source solutions without vendor dependencies. ### [hermes-mcp](https://mcpcatalogs.com/en/server/cloudwalk-hermes-mcp) by cloudwalk · score 50 · ★368 A high-performance Elixir SDK providing complete MCP client and server implementations with Elixir's concurrency and fault tolerance. **When to choose:** Teams developing Elixir/Phoenix applications who need MCP server/client functionality with Elixir's concurrency benefits. **When NOT to choose:** Projects not using Elixir, as this SDK is specific to the Elixir ecosystem and requires familiarity with BEAM. ### [abcoder](https://mcpcatalogs.com/en/server/cloudwego-abcoder) by cloudwego · score 50 · ★370 ABCoder is an AI-oriented code-processing framework that provides MCP tools for precise local code repository analysis. **When to choose:** Choose ABCoder when working with multiple programming languages and need deep, local code analysis without uploading proprietary code to external services. **When NOT to choose:** Not ideal for projects requiring real-time collaboration features or those already heavily invested in single-language analysis tools. ### [mcp-server-mas-sequential-thinking](https://mcpcatalogs.com/en/server/fradser-mcp-server-mas-sequential-thinking) by FradSer · score 50 · ★300 MCP server implementing a Multi-Agent System for sequential thinking with 6 specialized AI agents. **When to choose:** Choose this when you need deep, multi-perspective analysis of complex problems in your LLM workflows and have sufficient token budget. **When NOT to choose:** Avoid if you're token-constrained or need simple one-shot processing, as this system consumes 5-10x more tokens than simpler alternatives. ### [yantrikdb-server](https://mcpcatalogs.com/en/server/yantrikos-yantrikdb-server) by yantrikos · score 50 · ★143 Cognitive memory database for AI agents that consolidates duplicates, detects contradictions, and fades stale memories via temporal decay. **When to choose:** Choose YantrikDB when building AI agents that need sophisticated memory management beyond simple retrieval, especially when dealing with long-term memory, conflict detection, or temporal relevance. **When NOT to choose:** Not ideal if you need simple key-value storage without cognitive features; AGPL license may be problematic for commercial use without compliance. ### [A2V](https://mcpcatalogs.com/en/server/devin-axis-a2v) by Devin-AXIS · score 50 · ★1,200 A2V Protocol is an MCP-compatible AI value computation protocol that enables measurement, exchange, and settlement of AI agent value on blockchain. **When to choose:** Choose A2V when you need to build an economic layer for AI agents with blockchain-based value measurement and compensation. **When NOT to choose:** Don't choose A2V if you're looking for an MCP server to extend your AI agent's capabilities, as this is primarily a blockchain protocol with MCP compatibility claims. ### [STS2MCP](https://mcpcatalogs.com/en/server/gennadiyev-sts2mcp) by Gennadiyev · score 50 · ★344 MCP server for Slay the Spire 2 that exposes game state and actions for AI agents. **When to choose:** Choose this MCP server if you're researching AI decision-making in game environments or want AI agents to play Slay the Spire 2 autonomously with full game state access. **When NOT to choose:** Don't choose if you need support for games other than Slay the Spire 2, require multiplayer AI support for games besides STS2, or prefer solutions without external game modding requirements. ### [Frappe_Assistant_Core](https://mcpcatalogs.com/en/server/buildswithpaul-frappe-assistant-core) by buildswithpaul · score 50 · ★229 MCP server connecting LLMs to ERPNext with 24 tools for document management, search, reports, and workflows. **When to choose:** Choose Frappe Assistant Core if you're already using ERPNext and want to integrate AI capabilities directly into your business workflows without data migration. **When NOT to choose:** Avoid this if you need a solution for non-Frappe ERP systems or if you require strict on-premises AI model deployment without cloud dependencies. ### [EDDI](https://mcpcatalogs.com/en/server/labsai-eddi) by labsai · score 50 · ★353 EDDI is a Java-based multi-agent orchestration middleware with native MCP support for production-grade AI systems. **When to choose:** Choose EDDI for production-grade multi-agent orchestration when you need Java-based deployment, enterprise compliance features, or integration with existing Java/Quarkus ecosystems. **When NOT to choose:** Avoid EDDI for lightweight prototyping or if you prefer Python-based multi-agent frameworks like LangGraph or CrewAI. ### [opik-mcp](https://mcpcatalogs.com/en/server/comet-ml-opik-mcp) by comet-ml · score 50 · ★203 Comet-ML's MCP server for Opik provides unified access to prompts, projects, traces, and metrics across various IDEs. **When to choose:** Choose this MCP server if you're already using or planning to use Opik for MLOps and want seamless IDE integration with prompts, projects, traces, and metrics. **When NOT to choose:** Avoid if you need vendor-agnostic solutions, as this locks you into Comet's Opik platform, or if you require a hosted remote MCP service (currently not provided). ### [comfyui-mcp-server](https://mcpcatalogs.com/en/server/joenorton-comfyui-mcp-server) by joenorton · score 50 · ★313 Lightweight Python MCP server for local ComfyUI that enables AI agents to generate and iteratively refine images, audio, and video. **When to choose:** Choose when working with ComfyUI and want AI agents to generate/refine media assets through natural conversation. **When NOT to choose:** Don't choose if you need cloud-based AI generation, want to avoid local setup complexity, or require multi-user collaboration features. ### [deepseek-mcp-server](https://mcpcatalogs.com/en/server/dmontgomery40-deepseek-mcp-server) by DMontgomery40 · score 50 · ★334 A comprehensive MCP server exposing DeepSeek V4 API tools for chat completions, model listing, and balance checking. **When to choose:** Choose this MCP server if you need to integrate DeepSeek's language models into your workflow via MCP protocol, especially if you want access to their thinking and reasoning features. **When NOT to choose:** Don't choose this if you need features not exposed by DeepSeek's V4 API, such as speculative image/video upload tools or V4 monitoring capabilities. ### [studio](https://mcpcatalogs.com/en/server/decocms-studio) by decocms · score 50 · ★371 An open-source control plane for AI agents, integrating MCP tools with tracking and organization. **When to choose:** Choose Deco CMS when you need a unified control plane for managing multiple AI agents and MCP tools with project-based organization, cost tracking, and adaptive UIs. **When NOT to choose:** Avoid Deco CMS if you need simple point-to-point MCP integration without the overhead of a control plane, or if you're building revenue-generating SaaS (requires commercial license). ### [openclaw-mcp](https://mcpcatalogs.com/en/server/freema-openclaw-mcp) by freema · score 50 · ★169 MCP server bridging Claude.ai with self-hosted OpenClaw assistant using OAuth2 authentication. **When to choose:** Choose this server when you need secure integration between Claude.ai and a self-hosted OpenClaw assistant with proper authentication and multi-instance management. **When NOT to choose:** Don't choose this if you're not using OpenClaw or if you need a solution without OAuth2 authentication requirements. ### [scholar-search-mcp](https://mcpcatalogs.com/en/server/silung-scholar-search-mcp) by Silung · score 50 · ★180 MCP server for academic paper search integrating Semantic Scholar and arXiv with practical research tools. **When to choose:** Choose Scholar Search MCP when you need unified academic search across multiple scholarly sources for AI assistants, especially when working on literature reviews or research papers. **When NOT to choose:** Don't choose this if you need write access to academic databases or require access to non-academic research sources. ### [ClawMem](https://mcpcatalogs.com/en/server/yoloshii-clawmem) by yoloshii · score 50 · ★168 On-device memory layer for AI agents with MCP server integration and hybrid RAG search. **When to choose:** Choose ClawMem if you need a local, persistent memory layer for AI coding agents that maintains context across sessions and works offline without cloud dependencies. **When NOT to choose:** Don't choose ClawMem if you prefer a cloud-based memory system or need integration with non-supported AI agents that don't have MCP plugins or hooks. ### [squeez](https://mcpcatalogs.com/en/server/claudioemmanuel-squeez) by claudioemmanuel · score 49 · ★121 squeez is an MCP server that provides token compression and optimization for AI CLI hosts via hooks. **When to choose:** Choose Squeez when working with multiple AI CLI hosts and need significant token compression, particularly for bash-heavy workflows. **When NOT to choose:** Avoid Squeez if you primarily use AI assistants that don't support hook-based integration or if you need write capabilities beyond compression. ### [mcp-server-typescript](https://mcpcatalogs.com/en/server/dataforseo-mcp-server-typescript) by dataforseo · score 49 · ★198 A comprehensive MCP server for DataForSEO APIs that provides SEO data, keyword research, SERP tracking, and domain analytics. **When to choose:** Choose this MCP server when you need comprehensive SEO data access through AI assistants and already have or are willing to set up DataForSEO API credentials. **When NOT to choose:** Avoid this if you need free SEO data access (requires paid DataForSEO subscription) or if you need integration with non-SEO APIs. ### [memory-bank-mcp](https://mcpcatalogs.com/en/server/alioshr-memory-bank-mcp) by alioshr · score 49 · ★904 A TypeScript MCP server for centralized remote memory bank management across multiple projects. **When to choose:** Choose this when you need multi-project memory management with persistent context across development sessions, especially when working with AI assistants like Claude or Cline. **When NOT to choose:** Don't choose this if you need direct filesystem access beyond memory banking operations or require complex metadata management beyond simple file structure. ### [remindb](https://mcpcatalogs.com/en/server/radimsem-remindb) by radimsem · score 49 · ★93 Remindb is an agentic memory database that cuts session tokens by 75–99% using a portable SQLite file with MCP tool integration. **When to choose:** Choose remindb when you need a portable, efficient memory system for AI agents that handle large volumes of documentation or notes, especially when token costs are a concern. **When NOT to choose:** Don't choose remindb if you need real-time collaboration features or write access to the underlying database, as it's primarily designed for read-heavy memory scenarios. ### [pctx](https://mcpcatalogs.com/en/server/portofcontext-pctx) by portofcontext · score 49 · ★252 pctx is a Rust-based framework that connects AI agents to tools and MCP servers through Code Mode for secure, token-efficient workflows. **When to choose:** Choose pctx when you need to connect AI agents to multiple MCP services through a unified interface and want to optimize token usage by executing code in a sandbox rather than making sequential tool calls. **When NOT to choose:** Avoid pctx if you need fine-grained control over tool execution, have very specific tool requirements that can't be expressed through code, or require custom sandboxing beyond what Deno provides. ### [computer-use-mcp](https://mcpcatalogs.com/en/server/domdomegg-computer-use-mcp) by domdomegg · score 49 · ★256 An MCP server that gives Claude AI complete control of your computer for desktop automation tasks. **When to choose:** Choose this when you need an AI agent to interact with desktop applications, automate desktop tasks, or control legacy software that lacks APIs. **When NOT to choose:** Don't choose this for production systems or sensitive data - it gives complete control of your computer with no safety boundaries beyond user permissions. ### [clawfu-skills](https://mcpcatalogs.com/en/server/guia-matthieu-clawfu-skills) by guia-matthieu · score 49 · ★108 A curated library of 175 expert marketing methodologies for AI agents, accessible via MCP. **When to choose:** Choose ClawFu Skills when you need access to validated marketing expertise for AI agents, particularly for content creation, sales strategies, and business positioning frameworks. **When NOT to choose:** Not suitable if you need domain expertise outside of marketing, or if you require customizable, executable code rather than structured instruction sets. ### [LycheeMem](https://mcpcatalogs.com/en/server/lycheemem-lycheemem) by LycheeMem · score 49 · ★234 LycheeMemory is a lightweight long-term memory framework for LLM agents that supports MCP as an HTTP server. **When to choose:** Choose LycheeMemory when you need lightweight, structured long-term memory for LLM agents with automatic consolidation across multiple agent runtimes. **When NOT to choose:** Avoid LycheeMemory if you need a graph database solution as it uses SQLite + LanceDB instead of Neo4j, or if you require memory systems with complex multi-agent coordination features. ### [claude-prompts](https://mcpcatalogs.com/en/server/minipuft-claude-prompts) by minipuft · score 49 · ★148 MCP prompt template server offering hot-reloadable prompts, validation gates, and workflow composition for various AI clients. **When to choose:** Choose this if you need versioned prompt templates with validation gates and want to compose complex multi-step AI workflows that thread context between steps. **When NOT to choose:** Avoid if you need a simple one-shot prompt solution without workflow capabilities or if you prefer open-source alternatives without potential vendor lock-in. ### [gemini-flow](https://mcpcatalogs.com/en/server/clduab11-gemini-flow) by clduab11 · score 49 · ★383 Gemini-Flow is an archived production-ready AI orchestration platform with dual A2A and MCP protocol support. **When to choose:** Not applicable as this project is archived and no longer maintained. **When NOT to choose:** This project has been archived and is no longer maintained, making it unsuitable for any production use. ### [volcano-agent-sdk](https://mcpcatalogs.com/en/server/kong-volcano-agent-sdk) by Kong · score 49 · ★392 TypeScript SDK for building AI agents that combine LLM reasoning with MCP tools, supporting multiple providers and advanced patterns. **When to choose:** Choose this SDK when building complex multi-agent workflows with specialized roles that need to coordinate through MCP tools, especially when working with multiple LLM providers in a single application. **When NOT to choose:** Avoid this if you need a simpler agent framework without multi-agent coordination capabilities or if you're looking for a solution with extensive built-in tools rather than MCP integration. ### [ogham-mcp](https://mcpcatalogs.com/en/server/ogham-mcp-ogham-mcp) by ogham-mcp · score 49 · ★104 Ogham MCP is a persistent, searchable shared memory system for AI coding agents across clients. **When to choose:** Choose Ogham when you need persistent, cross-client memory for AI coding agents and already use or are willing to set up PostgreSQL/Supabase infrastructure. **When NOT to choose:** Don't choose Ogham if you need a fully self-contained solution without external database dependencies, or if you prefer simpler tools with less configuration overhead. ### [persistent-ai-memory](https://mcpcatalogs.com/en/server/savantskie-persistent-ai-memory) by savantskie · score 49 · ★230 Persistent AI memory system for LLMs with MCP server integration, database storage, and conversation tracking. **When to choose:** Choose this when you need persistent memory for AI assistants across different platforms, especially if you're already using OpenWebUI or want a local-first solution. **When NOT to choose:** Avoid if you need cloud-based memory synchronization or if you require a more comprehensive knowledge management system with advanced graph relationships. ### [z-image-studio](https://mcpcatalogs.com/en/server/iconben-z-image-studio) by iconben · score 49 · ★115 A comprehensive MCP server for Z-Image-Turbo text-to-image generation with CLI, Web UI, and multiple GPU optimizations. **When to choose:** Choose this MCP server when you need local image generation with support for multiple hardware platforms and want both CLI and web interfaces alongside MCP integration. **When NOT to choose:** Don't choose this if you need cloud-based image generation without local hardware requirements or if you need integration with image generation models other than Z-Image-Turbo. ### [paiml-mcp-agent-toolkit](https://mcpcatalogs.com/en/server/paiml-paiml-mcp-agent-toolkit) by paiml · score 49 · ★153 PMAT is an MCP server toolkit providing deterministic code analysis and AI context generation for agents. **When to choose:** Choose PMAT when you need comprehensive code analysis and quality enforcement for multi-language projects, especially when integrating with Claude Code or other AI agents for deterministic development workflows. **When NOT to choose:** Avoid PMAT if you need simple, lightweight code analysis tools or if you're working with a monolingual codebase where its extensive feature set would be overkill. ### [FirstData](https://mcpcatalogs.com/en/server/mlt-oss-firstdata) by MLT-OSS · score 49 · ★161 Comprehensive external fact context layer for AI agents with MCP server integration for accessing authoritative structured data sources. **When to choose:** Choose FirstData when you need verified, authoritative data sources for research or analysis and want guidance on navigating complex official websites. **When NOT to choose:** Don't choose FirstData if you need real-time data updates, custom data processing, or if you require access to non-authoritative or privately held datasets. ### [pubmed-mcp-server](https://mcpcatalogs.com/en/server/cyanheads-pubmed-mcp-server) by cyanheads · score 49 · ★97 A comprehensive MCP server providing access to PubMed and NCBI E-utilities with 9 tools for research and citation management. **When to choose:** Choose this server for biomedical research and academic applications when you need comprehensive PubMed integration with citation generation, full-text retrieval, and MeSH term exploration capabilities. **When NOT to choose:** Not suitable if you need access to non-biomedical research databases, as this server is specifically designed for PubMed/NCBI content only. ### [omem](https://mcpcatalogs.com/en/server/ourmem-omem) by ourmem · score 49 · ★196 Persistent memory system for AI agents with MCP Server plugin enabling cross-agent knowledge sharing. **When to choose:** Choose OMEM when building AI applications that need persistent memory across sessions and tools, or when multiple AI agents need to share knowledge with provenance tracking. **When NOT to choose:** Avoid OMEM if you need an open-source solution without vendor lock-in, or if you're looking for a simple memory system without the complexity of multi-tier sharing and reconciliation. ### [mem0-mcp](https://mcpcatalogs.com/en/server/pinkpixel-dev-mem0-mcp) by pinkpixel-dev · score 49 · ★95 An MCP server providing persistent memory for AI agents using Mem0.ai with cloud, Supabase, or local storage options. **When to choose:** Choose mem0-mcp when you need persistent memory for AI agents without building a custom memory system, especially if you're already using or comfortable with Supabase or want cloud-hosted memory solutions. **When NOT to choose:** Avoid mem0-mcp if you need full control over your data storage infrastructure, have strict data residency requirements, or prefer a self-contained solution without external API dependencies. ### [neurolink](https://mcpcatalogs.com/en/server/juspay-neurolink) by juspay · score 49 · ★90 NeuroLink is a universal AI integration platform that unifies 21+ AI providers and 100+ models under one consistent API with MCP server support. **When to choose:** Choose NeuroLink when you need to integrate multiple AI providers under a single API with enterprise-grade features like multi-provider failover and intelligent routing. **When NOT to choose:** Avoid NeuroLink if you need a lightweight solution for a single AI provider or prefer a simpler integration without the complexity of enterprise features. ### [AgentCrew](https://mcpcatalogs.com/en/server/saigontechnology-agentcrew) by saigontechnology · score 49 · ★200 AgentCrew is a Python-based multi-agent system that supports MCP for specialized AI collaboration. **When to choose:** Choose AgentCrew when you need a complete multi-agent chat application with built-in UI and workflows that also supports MCP for external tool integration. **When NOT to choose:** Don't choose AgentCrew if you need a dedicated MCP server for integration into existing AI tooling, as it's primarily a chat application with MCP support rather than a standalone MCP implementation. ### [prism-coder](https://mcpcatalogs.com/en/server/dcostenco-prism-coder) by dcostenco · score 49 · ★133 MCP server providing persistent memory, semantic search, and local LLM for AI agents. **When to choose:** Choose Prism Coder when you need AI agents with persistent memory across sessions and require offline functionality without API dependencies. **When NOT to choose:** Avoid if you need cloud-based AI processing, require real-time collaboration features, or need extensive write capabilities beyond its memory system. ### [notebooklm-skill](https://mcpcatalogs.com/en/server/claude-world-notebooklm-skill) by claude-world · score 49 · ★209 MCP server connecting NotebookLM research with Claude content creation across multiple platforms. **When to choose:** Choose this when you need to combine NotebookLM's research capabilities with Claude's content generation for structured research-to-content workflows. **When NOT to choose:** Don't choose this if you need offline access or want to avoid Google account dependency for authentication. ### [mengram](https://mcpcatalogs.com/en/server/alibaizhanov-mengram) by alibaizhanov · score 48 · ★167 Mengram is a human-like memory system for AI agents with semantic, episodic & procedural memory that learns from failures. **When to choose:** Choose Mengram for AI agents that need persistent, evolving memory systems capable of learning from failures, especially when working with multi-language contexts or procedural knowledge. **When NOT to choose:** Avoid Mengram if you need a fully self-hosted solution, as it requires cloud API access, or if you only need simple fact storage without procedural memory capabilities. ### [medical-mcp](https://mcpcatalogs.com/en/server/jamesanz-medical-mcp) by JamesANZ · score 48 · ★90 Medical MCP provides trusted medical data from FDA, WHO, PubMed, RxNorm, and Google Scholar with no API keys required. **When to choose:** Choose this server when you need reliable medical information in your AI workflow without requiring API keys or compromising privacy. **When NOT to choose:** Don't choose this for clinical decision-making as it's intended for informational purposes only and should not replace professional medical advice. ### [mcp-agent-langchainjs](https://mcpcatalogs.com/en/server/azure-samples-mcp-agent-langchainjs) by Azure-Samples · score 48 · ★183 A burger ordering AI agent system using LangChain.js and MCP servers to interact with a restaurant API. **When to choose:** Teams looking to build production-ready MCP servers for e-commerce scenarios, especially those already using Azure serverless infrastructure. **When NOT to choose:** For non-Azure environments or if you need a simpler MCP implementation without the complexity of a full multi-service architecture. ### [ClaudeHistoryMCP](https://mcpcatalogs.com/en/server/jhammant-claudehistorymcp) by jhammant · score 48 · ★65 MCP server for searching Claude Code conversation history with hybrid BM25 + TF-IDF search and knowledge extraction. **When to choose:** Choose ClaudeHistoryMCP when you work frequently with Claude Code and need to efficiently search and retrieve past solutions, decisions, and patterns from your conversation history. **When NOT to choose:** Avoid this if you're not using Claude Code, as it's specifically designed to work with Claude's conversation format and stores data in Claude's directory structure. ### [UnrealGenAISupport](https://mcpcatalogs.com/en/server/prajwalshettydev-unrealgenaisupport) by prajwalshettydev · score 48 · ★587 Unreal Engine plugin supporting 200+ AI models with MCP server functionality for scene control and blueprint generation. **When to choose:** Choose this Unreal Engine plugin if you're already using Unreal for game development and want to integrate AI models for NPC behavior, scene generation, and blueprint automation through MCP protocol. **When NOT to choose:** Avoid this if you need write access to project files, require a more stable alternative with automated testing, or are looking for official Unreal MCP integration planned for UE 5.8+. ### [notebooklm-mcp-secure](https://mcpcatalogs.com/en/server/pantheon-security-notebooklm-mcp-secure) by Pantheon-Security · score 48 · ★63 Enterprise-grade MCP server for Google NotebookLM with 17 security layers and advanced features like Gemini Deep Research. **When to choose:** Choose this server for enterprise deployments requiring NotebookLM integration with comprehensive security, audit trails, and compliance documentation. **When NOT to choose:** Not suitable for simple use cases or environments where the additional security overhead is unnecessary; consider the original notebooklm-mcp for basic functionality. ### [Zikkaron](https://mcpcatalogs.com/en/server/amanhij-zikkaron) by amanhij · score 48 · ★59 Biologically-inspired persistent memory engine for Claude Code with 26 cognitive subsystems, local SQLite storage, and advanced memory recall capabilities. **When to choose:** Choose Zikkaron when working on long-term development projects where you need context continuity across sessions and want to avoid repeating explanations of architecture decisions or debugging approaches. **When NOT to choose:** Don't choose Zikkaron if you need memory systems with strong abstention capabilities or if you're working on projects where precise chronological ordering of events is critical. ### [penpot-mcp](https://mcpcatalogs.com/en/server/montevive-penpot-mcp) by montevive · score 48 · ★228 Penpot MCP server integrates AI assistants like Claude with Penpot designs for automated design workflows and analysis. **When to choose:** Choose this if you're working with Penpot designs and want AI assistance for design analysis, documentation, and workflow automation. **When NOT to choose:** Avoid if you need write access to design files (this server appears read-only) or if you're not using Penpot as your design platform. ### [houtini-lm](https://mcpcatalogs.com/en/server/houtini-ai-houtini-lm) by houtini-ai · score 48 · ★88 MCP server that offloads bounded tasks from Claude Code to local/cloud LLMs to reduce token consumption. **When to choose:** Developers already using Claude Code who want to reduce API costs by offloading bounded tasks to local or cloud LLMs without compromising on complex reasoning. **When NOT to choose:** Users who need ultra-low latency responses, as local LLM inference is typically 3-30× slower than frontier models. ### [comfy-pilot](https://mcpcatalogs.com/en/server/constantineb6-comfy-pilot) by ConstantineB6 · score 48 · ★186 MCP server + embedded terminal allowing Claude Code to view, edit, and run ComfyUI workflows programmatically. **When to choose:** Choose Comfy Pilot when working with ComfyUI workflows and wanting to leverage natural language to create, modify, and troubleshoot complex image generation pipelines without manual node manipulation. **When NOT to choose:** Avoid if you need workflow portability across different AI image generation tools beyond ComfyUI, or if you prefer a visual programming approach without AI assistance. ### [remote-mcp-server](https://mcpcatalogs.com/en/server/gleanwork-remote-mcp-server) by gleanwork · score 48 · ★160 Enterprise MCP server providing secure access to company knowledge, documents, and expertise. **When to choose:** Choose this MCP server if you need secure, permission-aware AI access to your organization's internal knowledge base and documents through existing MCP-compatible tools. **When NOT to choose:** Avoid this if you're not already using Glean as your enterprise knowledge platform, as it requires integration with their specific system. ### [math-mcp](https://mcpcatalogs.com/en/server/ethanhenrickson-math-mcp) by EthanHenrickson · score 48 · ★133 MCP server providing math, statistical and trigonometric functions for accurate numerical calculations. **When to choose:** Choose Math-MCP when your application requires accurate mathematical calculations and statistical operations that LLMs might otherwise approximate incorrectly. **When NOT to choose:** Don't choose Math-MCP if you need advanced mathematical operations like matrix algebra, calculus, or complex number computations beyond its scope. ### [Matryoshka](https://mcpcatalogs.com/en/server/yogthos-matryoshka) by yogthos · score 48 · ★134 MCP server for analyzing documents larger than context windows using recursive language models and symbolic commands. **When to choose:** Choose Matryoshka when you need to analyze very large documents that exceed your LLM's context window, especially when traditional chunking or vector database approaches lose important connections across document sections. **When NOT to choose:** Don't choose Matryoshka if you need fine-grained control over individual document chunks or require advanced metadata preservation that goes beyond its symbolic language capabilities. ### [mcp-echarts](https://mcpcatalogs.com/en/server/hustcc-mcp-echarts) by hustcc · score 48 · ★228 MCP ECharts enables dynamic chart generation using ECharts through the Model Context Protocol. **When to choose:** Choose mcp-echarts when you need to generate dynamic visualizations from AI assistants with ECharts' comprehensive charting capabilities and want local processing without external dependencies. **When NOT to choose:** Avoid if you need specialized chart types not supported by ECharts, or if you require cloud-based rendering rather than local processing. ### [arxiv-latex-mcp](https://mcpcatalogs.com/en/server/takashiishida-arxiv-latex-mcp) by takashiishida · score 48 · ★131 MCP server enabling LLMs to access and process arXiv papers via original LaTeX source for better math handling. **When to choose:** Choose this server when working with mathematical papers where precise equation interpretation is critical, especially in computer science, mathematics, and engineering fields. **When NOT to choose:** Avoid this server if you primarily need to analyze non-technical papers or those without mathematical content, or if you require processing capabilities beyond LaTeX source extraction. ### [langfuse-mcp](https://mcpcatalogs.com/en/server/avivsinai-langfuse-mcp) by avivsinai · score 48 · ★87 Langfuse MCP server for AI agents to query trace data, debug exceptions, and manage observability. **When to choose:** Choose this MCP server if you need comprehensive access to Langfuse observability data including traces, sessions, exceptions, and dataset management, especially when the official server's limited functionality isn't sufficient. **When NOT to choose:** Don't choose this if you only need prompt management capabilities (covered by the official server), or if you're concerned about potential vendor lock-in with Langfuse-specific features. ### [omega-memory](https://mcpcatalogs.com/en/server/omega-memory-omega-memory) by omega-memory · score 48 · ★138 Cross-model memory system for AI agents that persists across sessions with MCP server support. **When to choose:** Choose OMEGA when you need persistent memory for AI coding assistants across sessions and want to avoid cloud dependencies or vendor lock-in. **When NOT to choose:** Don't choose OMEGA if you need cloud-based memory synchronization across devices or require extensive multi-agent coordination features (only available in the paid pro version). ### [vesta-mac-dist](https://mcpcatalogs.com/en/server/scouzi1966-vesta-mac-dist) by scouzi1966 · score 48 · ★80 Vesta is a multi-backend AI chat app for macOS with full MCP server integration for programmatic control. **When to choose:** Choose Vesta when you need comprehensive programmatic control over multiple AI backends on macOS, especially if you're already using Claude Code. **When NOT to choose:** Avoid Vesta if you're not using macOS or Apple Silicon hardware, or if you prefer cloud-based AI solutions without local model support. ### [exstruct](https://mcpcatalogs.com/en/server/harumiweb-exstruct) by harumiWeb · score 48 · ★144 ExStruct is a Python package that converts Excel files to structured JSON with MCP integration for AI agents. **When to choose:** Choose ExStruct when working with Excel data in LLM/RAG pipelines or when AI agents need to read/write Excel files with structured output. **When NOT to choose:** Avoid if you need complex Excel macros support or advanced formatting that isn't preserved in the JSON extraction process. ### [agentor](https://mcpcatalogs.com/en/server/celestoai-agentor) by CelestoAI · score 48 · ★180 Agentor is a comprehensive framework for building and deploying AI agents with MCP server capabilities, tools, and agent-to-agent communication. **When to choose:** Choose Agentor if you need a production-ready MCP server that integrates directly with FastAPI applications and want to deploy AI agents with skills and tools to a scalable platform. **When NOT to choose:** Avoid Agentor if you need a standalone MCP server without FastAPI integration or if you prefer an open-source deployment solution without vendor lock-in to Celesto's platform. ### [TrueMemory](https://mcpcatalogs.com/en/server/buildingjoshbetter-truememory) by buildingjoshbetter · score 48 · ★77 A high-performance memory system for AI agents that runs locally on SQLite with MCP server integration. **When to choose:** Choose TrueMemory when you need persistent memory for AI agents that works offline and prioritizes privacy by keeping all data locally. **When NOT to choose:** Don't choose this if you need true MCP server functionality for custom tool integration - it's primarily a memory system with MCP hooks, not a standalone MCP server. ### [selene](https://mcpcatalogs.com/en/server/tercumantanumut-selene) by tercumantanumut · score 48 · ★166 Selene is an AI agent desktop app that integrates MCP tools for WhatsApp, Telegram, Slack, and Discord. **When to choose:** Choose Selene if you want a comprehensive desktop application for AI agents that includes MCP integration rather than a focused MCP server for a specific functionality. **When NOT to choose:** Don't choose Selene if you need a dedicated MCP server for a specific integration or if you prefer smaller, focused tools rather than an all-in-one AI platform. ### [MakerAi](https://mcpcatalogs.com/en/server/gustavoeenriquez-makerai) by gustavoeenriquez · score 48 · ★173 MakerAI is a complete AI ecosystem for Delphi with MCP server/client support, RAG, and autonomous agents. **When to choose:** Choose MakerAI if you're building Delphi applications that need integrated AI capabilities, autonomous agents, or RAG systems with native Pascal code and full API access. **When NOT to choose:** Don't choose MakerAI if you're not using Delphi or Pascal, or if you need a solution that supports platforms beyond Windows and macOS (Delphi native limitations). ### [sympy-mcp](https://mcpcatalogs.com/en/server/sdiehl-sympy-mcp) by sdiehl · score 48 · ★70 A Model Context Protocol server that exposes SymPy's symbolic mathematics capabilities through MCP tool calling for LLMs. **When to choose:** Choose this server when you need precise symbolic mathematics calculations and want to leverage LLMs for orchestrating mathematical workflows without risking hallucinations in mathematical operations. **When NOT to choose:** Don't choose this server if you need numerical computation rather than symbolic manipulation, or if you're working in an environment where Python/uv package management is not feasible. ### [rustchain-mcp](https://mcpcatalogs.com/en/server/scottcjn-rustchain-mcp) by Scottcjn · score 48 · ★65 MCP server connecting AI agents to RustChain blockchain, BoTTube video platform, and Beacon messaging. **When to choose:** Choose this server if your AI agent needs to interact with blockchain services, participate in a decentralized AI economy, or communicate with other agents through the Beacon protocol. **When NOT to choose:** Don't choose this if you need a general-purpose blockchain integration or if you require support for other blockchain networks beyond RustChain. ### [welink](https://mcpcatalogs.com/en/server/runzhliu-welink) by runzhliu · score 48 · ★142 WeLink is an AI-powered platform for analyzing WeChat chat data, featuring MCP server integration for direct querying in Claude Code. **When to choose:** Choose WeLink MCP server if you want to query your WeChat chat data directly from Claude Code without using the web interface, especially if you already use WeLink for chat analysis. **When NOT to choose:** Don't choose this MCP server if you don't use WeChat or haven't set up WeLink to process your chat data first, as the MCP server depends on the local WeLink database. ### [hayhooks](https://mcpcatalogs.com/en/server/deepset-ai-hayhooks) by deepset-ai · score 48 · ★141 Hayhooks deploys Haystack pipelines as REST APIs and MCP tools for AI environments. **When to choose:** Choose Hayhooks when you need to deploy Haystack pipelines as MCP tools in AI development environments or as chat interfaces with minimal setup. **When NOT to choose:** Avoid Hayhooks if you need fine-grained authentication controls or are not already using Haystack for your pipelines, as it's specifically designed for Haystack integration. ### [notebooklm-mcp](https://mcpcatalogs.com/en/server/roomi-fields-notebooklm-mcp) by roomi-fields · score 48 · ★76 Google NotebookLM MCP server with HTTP API for Q&A, content generation, and multi-account rotation. **When to choose:** Choose this when you need both MCP integration for AI assistants and HTTP API for workflow automation, with support for full Studio content generation beyond simple Q&A. **When NOT to choose:** Avoid if you need only MCP protocol without HTTP API, or if you require content generation features not available in NotebookLM's Studio. ### [mem0-mcp-selfhosted](https://mcpcatalogs.com/en/server/elvismdev-mem0-mcp-selfhosted) by elvismdev · score 48 · ★84 A self-hosted mem0 MCP server that integrates with Qdrant, Neo4j, and Ollama for persistent memory management with Claude Code. **When to choose:** Choose this when you need persistent, cross-session memory for Claude Code with full control over your data infrastructure. **When NOT to choose:** Avoid if you want a cloud-based solution without managing your own vector database and graph storage infrastructure. ### [academia_mcp](https://mcpcatalogs.com/en/server/ilyagusev-academia-mcp) by IlyaGusev · score 47 · ★89 Academia MCP server provides tools for searching, fetching, and analyzing scientific papers and research datasets. **When to choose:** Choose this server if you need comprehensive academic research tools and are already using or can integrate Python-based LLM services into your workflow. **When NOT to choose:** Avoid this if you need a purely non-Python solution, require proprietary document formats support beyond PDF/LaTeX, or need high-volume commercial use without proper licensing review. ### [roampal](https://mcpcatalogs.com/en/server/roampal-ai-roampal) by roampal-ai · score 47 · ★120 Roampal is an AI memory system with MCP integration that learns from user feedback across conversations. **When to choose:** Choose Roampal when you need a privacy-first AI memory system that learns from your feedback and works entirely offline with local models. **When NOT to choose:** Avoid if you need cloud-based AI services, real-time collaboration features, or require multi-user support. ### [vectorize-mcp-server](https://mcpcatalogs.com/en/server/vectorize-io-vectorize-mcp-server) by vectorize-io · score 47 · ★106 Vectorize MCP server enables vector search, text extraction, and deep research through the Model Context Protocol. **When to choose:** Choose this MCP server if you're already using Vectorize or need vector database capabilities with document extraction and research tools. **When NOT to choose:** Avoid if you need open-source vector databases without vendor lock-in or if you don't want to rely on external API services. ### [pdf-mcp](https://mcpcatalogs.com/en/server/jztan-pdf-mcp) by jztan · score 47 · ★35 MCP server enabling AI agents to read, search, and extract content from large PDFs with chunked reading, hybrid search, OCR, and SQLite caching. **When to choose:** Choose pdf-mcp when you need comprehensive PDF processing capabilities for AI agents, especially for large documents requiring chunked reading, hybrid search, and OCR support. **When NOT to choose:** Don't choose pdf-mcp if you need encrypted PDF support, real-time collaborative PDF editing, or advanced multimedia processing beyond images. ### [imagesorcery-mcp](https://mcpcatalogs.com/en/server/sunriseapps-imagesorcery-mcp) by sunriseapps · score 47 · ★311 ImageSorcery MCP provides powerful locally-run image processing tools for AI assistants. **When to choose:** Choose this when you need privacy-focused local image processing with advanced computer vision capabilities and don't want to upload images to cloud services. **When NOT to choose:** Don't choose this if you need to process extremely large images or need specialized models not supported by the current implementation. ### [sparql-llm](https://mcpcatalogs.com/en/server/sib-swiss-sparql-llm) by sib-swiss · score 47 · ★108 MCP server that helps LLMs generate and validate SPARQL queries for biological knowledge graphs. **When to choose:** Choose this MCP server if you work with biological data endpoints and need LLM assistance generating validated SPARQL queries against knowledge graphs. **When NOT to choose:** Don't choose this if you're working with non-biological data endpoints, as it's specifically designed for SIB biodata resources like UniProt and Bgee. ### [openrouter-deep-research-mcp](https://mcpcatalogs.com/en/server/wheattoast11-openrouter-deep-research-mcp) by wheattoast11 · score 47 · ★50 Multi-agent research MCP server with async agents, semantic search, and persistent knowledge base. **When to choose:** Choose this MCP server when you need sophisticated multi-agent research capabilities with persistent knowledge storage and advanced semantic search features. **When NOT to choose:** Don't choose this if you need a lightweight research tool without the complexity of multiple agents or if you prefer open-source models over commercial APIs. ### [STAMP](https://mcpcatalogs.com/en/server/katherlab-stamp) by KatherLab · score 47 · ★119 STAMP is a computational pathology tool with MCP support for histopathology image analysis and biomarker prediction. **When to choose:** Pathology researchers needing a validated, peer-reviewed workflow for biomarker discovery from whole-slide images. **When NOT to choose:** If you need immediate MCP integration; STAMP only mentions MCP support as a future feature without implementation. ### [blender-open-mcp](https://mcpcatalogs.com/en/server/dhakalnirajan-blender-open-mcp) by dhakalnirajan · score 47 · ★97 An MCP server that connects Blender 3D with local AI models via Ollama for natural language control. **When to choose:** Choose this MCP server when you want to integrate local AI models into Blender workflows without cloud dependencies, specifically for natural language control of 3D creation. **When NOT to choose:** Avoid this if you need cloud-based AI models, require more advanced rendering capabilities, or prefer commercial 3D software like Maya with built-in AI features. ### [ChatSpatial](https://mcpcatalogs.com/en/server/cafferychen777-chatspatial) by cafferychen777 · score 47 · ★37 MCP server for spatial transcriptomics analysis through natural language interfaces. **When to choose:** Researchers working with spatial transcriptomics data who need reproducible, cross-platform analysis through natural language interfaces without writing custom scripts. **When NOT to choose:** Users needing specialized analysis methods not covered by the 15 available categories, or those requiring direct access to raw data manipulation beyond the schema-enforced orchestration. ### [Cortex](https://mcpcatalogs.com/en/server/cdeust-cortex) by cdeust · score 47 · ★34 Cortex is a neuroscience-based persistent memory system for Claude Code with 47 MCP tools that remembers context, decisions, and patterns across sessions. **When to choose:** Choose Cortex when you need long-term memory for code development across multiple sessions, especially when working on complex systems where context persistence significantly improves productivity. **When NOT to choose:** Avoid Cortex if you need a lightweight solution without PostgreSQL dependency, or if you're working on short-term projects where context persistence doesn't provide significant value. ### [mcp](https://mcpcatalogs.com/en/server/baidu-maps-mcp) by baidu-maps · score 47 · ★426 Baidu Map MCP Server provides comprehensive geospatial APIs for location-based services through MCP protocol. **When to choose:** Choose this MCP server when you need comprehensive geospatial data and location-based services in China or for applications requiring Baidu Maps' authoritative location data. **When NOT to choose:** Avoid if you need location services outside of China, prefer open-source map data, or require more advanced features not offered by Baidu Maps. ### [AgentUp](https://mcpcatalogs.com/en/server/always-further-agentup) by always-further · score 47 · ★116 AgentUp is a Python-based AI agent framework with built-in MCP support for secure, portable agent deployment. **When to choose:** Choose AgentUp when you need production-ready AI agents with security, consistent behavior across environments, and MCP integration capabilities. **When NOT to choose:** Avoid AgentUp if you need a simple chatbot without enterprise features or if you're looking for lightweight, minimalist agent frameworks. ### [openrouter-mcp-multimodal](https://mcpcatalogs.com/en/server/stabgan-openrouter-mcp-multimodal) by stabgan · score 47 · ★39 MCP server for OpenRouter providing text, image, audio, and video analysis and generation with 300+ LLMs. **When to choose:** Choose this server when you need multimodal capabilities (text, image, audio, video) in your AI workflows and want access to OpenRouter's diverse model ecosystem through MCP. **When NOT to choose:** Avoid if you need vendor-agnostic access or if you're locked into a different LLM provider that doesn't offer OpenRouter integration. ### [brainstorm-mcp](https://mcpcatalogs.com/en/server/spranab-brainstorm-mcp) by spranab · score 47 · ★57 MCP server for multi-model AI brainstorming debates with structured synthesis and cross-model code review. **When to choose:** When you need to compare perspectives from multiple AI models on technical decisions or code reviews, and want structured analysis of their disagreements. **When NOT to choose:** If you're looking for a single-model assistant or when your use case requires fast, single-model responses rather than multi-model debates. ### [fullstack-langgraph-nextjs-agent](https://mcpcatalogs.com/en/server/agentailor-fullstack-langgraph-nextjs-agent) by agentailor · score 47 · ★106 Production-ready Next.js template with MCP integration for dynamic tool loading in AI agents. **When to choose:** Choose this template when building production-ready LangGraph.js agents with MCP integration, especially if you need human-in-the-loop tool approval and persistent conversation memory with PostgreSQL. **When NOT to choose:** Don't choose this if you need a simple MCP server without the full Next.js stack, or if you prefer a different backend technology than Node.js with Prisma ORM. ### [ollama-mcp-bridge](https://mcpcatalogs.com/en/server/jonigl-ollama-mcp-bridge) by jonigl · score 47 · ★86 Ollama MCP Bridge provides an API layer that integrates multiple MCP servers with the Ollama API, enabling dynamic AI tool integration for local LLM applications. **When to choose:** Choose this when you want to extend local Ollama deployments with MCP tools without changing your existing application code. **When NOT to choose:** Not ideal if you need direct MCP integration without the Ollama layer or require write access to data sources (as most tools are read-only). ### [ollama-mcp](https://mcpcatalogs.com/en/server/rawveg-ollama-mcp) by rawveg · score 47 · ★160 TypeScript MCP server exposing complete Ollama SDK with 14 tools for local LLM integration. **When to choose:** Choose this MCP server if you want to integrate Ollama's local LLM capabilities with Claude Desktop or Cline, especially when you need both model operations and web search functionality through a single interface. **When NOT to choose:** Avoid this server if you're looking for open-source licensing (it's AGPL-3.0 which may have compliance implications) or if you need extensive tool calling capabilities beyond Ollama's native support. ### [grainulator](https://mcpcatalogs.com/en/server/grainulation-grainulator) by grainulation · score 47 · ★85 Multi-pass research tool with claim-based investigation and confidence scoring for Claude Code. **When to choose:** Choose Grainulator for research-intensive projects requiring structured evidence gathering, conflict detection, and confidence scoring before making decisions. **When NOT to choose:** Don't choose Grainulator if you need rapid, simple answers or prefer a simpler research process without the complexity of multiple investigation passes and claim tension analysis. ### [MCP-BOE](https://mcpcatalogs.com/en/server/computingvictor-mcp-boe) by ComputingVictor · score 47 · ★39 MCP server providing access to Spanish BOE legislation, daily summaries, and official documents via MCP and REST API. **When to choose:** Choose this server when you need comprehensive access to Spanish legal documents, legislation, and official publications in Spanish. **When NOT to choose:** Don't choose this if you need access to legal documents from countries other than Spain or if you require real-time updates beyond the BOE's publication schedule. ### [modular-context-obsidian-plugin](https://mcpcatalogs.com/en/server/klemensgc-modular-context-obsidian-plugin) by klemensgc · score 47 · ★88 An Obsidian plugin that provides MCP tools for Google Workspace and transforms your vault into an LLM-native knowledge base. **When to choose:** Choose this if you already use Obsidian as your knowledge base and want seamless integration with Google Workspace through Claude Code. **When NOT to choose:** Not suitable for non-macOS users as it's primarily tested on macOS, or if you need cross-platform compatibility with Google Workspace tools. ### [ckan-mcp-server](https://mcpcatalogs.com/en/server/ondata-ckan-mcp-server) by ondata · score 47 · ★46 MCP server for querying CKAN open data portals, enabling natural language access to datasets, organizations, and data. **When to choose:** Choose this when you need AI access to CKAN-based open data portals worldwide and want to search datasets, explore organizations, query tabular data, and read metadata through natural language. **When NOT to choose:** Don't choose this if you need access to non-CKAN open data platforms or require features beyond basic dataset search, metadata viewing, and simple SQL queries on tabular data. ### [memcord](https://mcpcatalogs.com/en/server/ukkit-memcord) by ukkit · score 47 · ★67 Privacy-first MCP server for AI memory management with intelligent summarization and search capabilities. **When to choose:** Choose Memcord when you need to preserve conversation context across multiple AI sessions with complete privacy control and local storage. **When NOT to choose:** Don't choose Memcord if you need cloud-based AI memory services or require team collaboration features across different users. ### [mcptube](https://mcpcatalogs.com/en/server/0xchamin-mcptube) by 0xchamin · score 47 · ★104 MCP server that transforms YouTube videos into a persistent knowledge base with transcripts, vision analysis, and agentic search. **When to choose:** Choose mcptube when you need to transform educational or tutorial videos into a searchable, compounding knowledge base that improves over time with more content. **When NOT to choose:** Avoid if you need real-time video analysis without ingestion delay, or if you're working with videos that don't benefit from scene-change detection. ### [mcp-agent](https://mcpcatalogs.com/en/server/haohao-end-mcp-agent) by Haohao-end · score 47 · ★85 A modular Python MCP framework integrating LLMs with tools, weather data, and advanced RAG capabilities. **When to choose:** Choose this Python-based MCP framework when building modular AI applications that require both real-time data retrieval and document-based knowledge augmentation. **When NOT to choose:** Not suitable for production systems requiring long-term persistence or those needing comprehensive authentication and authorization controls. ### [apple-health-mcp-server](https://mcpcatalogs.com/en/server/the-momentum-apple-health-mcp-server) by the-momentum · score 47 · ★189 MCP server for querying Apple Health data with natural language using DuckDB. **When to choose:** Choose this server if you want to analyze Apple Health data without SQL knowledge, using natural language queries with an LLM. **When NOT to choose:** Don't choose this if you need real-time health data sync or support for non-Apple health devices. ### [mcp-image](https://mcpcatalogs.com/en/server/shinpr-mcp-image) by shinpr · score 47 · ★110 MCP server for AI image generation with automatic prompt optimization using Gemini or OpenAI models. **When to choose:** Choose this server when you need AI-generated images without prompt engineering expertise, with automatic prompt enhancement and quality presets. **When NOT to choose:** Avoid this if you need maximum control over the exact prompt wording without any enhancement, or if you're looking for a wider variety of image generation models beyond Gemini and OpenAI. ### [fabric-mcp](https://mcpcatalogs.com/en/server/ksylvan-fabric-mcp) by ksylvan · score 47 · ★86 Fabric MCP Server integrates Daniel Miessler's Fabric AI framework into MCP-enabled tools through a standalone server. **When to choose:** Choose this MCP server when you're already using Fabric's prompt patterns and want to integrate them directly into MCP-compatible tools like IDEs and chat interfaces. **When NOT to choose:** Don't choose this if you're not already using Fabric, as it requires a separate Fabric instance running with the --serve flag. ### [Ombre-Brain](https://mcpcatalogs.com/en/server/p0lar1zzz-ombre-brain) by P0lar1zzZ · score 47 · ★94 An emotional memory system for Claude using Russell's valence/arousal coordinates with MCP protocol integration. **When to choose:** Choose this when you need Claude to maintain emotional context across conversations and want a human-like memory system with natural forgetting curves. **When NOT to choose:** Avoid if you need a simple key-value memory store or if you're not comfortable with a system that automatically forgets information over time. ### [reddit-research-mcp](https://mcpcatalogs.com/en/server/king-of-the-grackles-reddit-research-mcp) by king-of-the-grackles · score 47 · ★113 Reddit research MCP server providing competitive analysis, customer discovery and market insights with semantic search. **When to choose:** Choose this server when you need structured market research from Reddit with semantic search across thousands of communities and require evidence-based insights with full citations. **When NOT to choose:** Don't choose this if you need real-time access to very recent Reddit posts (limited to how often the server updates its index) or if you require processing private/subreddit-restricted content. ### [ask-user-questions-mcp](https://mcpcatalogs.com/en/server/paulp-o-ask-user-questions-mcp) by paulp-o · score 47 · ★108 AUQ is an MCP server enabling AI agents to ask clarifying questions via a separate CLI interface for human-in-the-loop workflows. **When to choose:** Choose AUQ when you need advanced human-in-the-loop capabilities for multi-agent workflows, especially when you want non-invasive CLI interaction with support for complex question types and remote usage. **When NOT to choose:** Don't choose AUQ if you need simpler, one-off questions without queuing capabilities, or if you prefer in-editor UI interaction rather than a separate CLI window for answering questions. ### [askimo](https://mcpcatalogs.com/en/server/askimo-ai-askimo) by askimo-ai · score 47 · ★110 A desktop-first AI agent platform that integrates MCP tools, enabling local-first workflows with multiple AI models. **When to choose:** Choose Askimo when you need a local-first AI platform that combines multiple AI providers with MCP tool integration, especially for RAG workflows and automation. **When NOT to choose:** Don't choose Askimo if you need cloud-based AI processing, require extensive customization of the MCP implementation, or need a web interface rather than a desktop/CLI application. ### [ainativelang](https://mcpcatalogs.com/en/server/sbhooley-ainativelang) by sbhooley · score 47 · ★87 AI Native Lang (AINL) is an MCP server that turns AI conversations into structured workflows with deterministic execution. **When to choose:** Choose AINL when building complex AI workflows that need repeatability, state management, and significant token savings, especially when working with multiple MCP-compatible AI agents. **When NOT to choose:** Don't choose AINL for simple, one-off tasks that don't require state or repeatability, or if you prefer working with natural language prompts over a structured programming approach. ### [iron-manus-mcp](https://mcpcatalogs.com/en/server/dnnyngyen-iron-manus-mcp) by dnnyngyen · score 46 · ★65 Model Context Protocol server for AI workflow orchestration with 8-phase agent delegation and todo-based task dispatch. **When to choose:** Choose this MCP server for complex AI workflows requiring specialized agents, structured planning phases, and secure API integration with SSRF protection. **When NOT to choose:** Don't choose this if you need a simple MCP server with basic tools, as this implements a complex multi-agent orchestration system with significant setup overhead. ### [Train-in-Silence](https://mcpcatalogs.com/en/server/hlpun-train-in-silence) by hlpun · score 46 · ★68 Task-aware MCP server that calculates optimal GPU configurations across 10+ cloud providers for LLM fine-tuning. **When to choose:** Choose this server when you need to optimize GPU resource selection for LLM fine-tuning across multiple cloud providers, balancing cost and performance. **When NOT to choose:** Don't choose this if you need direct GPU access control or have specific requirements not covered by the estimation model. ### [aura](https://mcpcatalogs.com/en/server/mezmo-aura) by mezmo · score 46 · ★63 A production-ready AI agent framework with MCP tool integration, RAG pipelines, and OpenAI-compatible API. **When to choose:** Choose Aura when you need to build production-ready multi-agent systems with tool calling, RAG integration, and an OpenAI-compatible API. **When NOT to choose:** Don't choose Aura if you need simple single-agent deployments, have limited Rust expertise, or require extensive client-side tooling without accepting security risks. ### [sample-agentic-ai-demos](https://mcpcatalogs.com/en/server/aws-samples-sample-agentic-ai-demos) by aws-samples · score 46 · ★101 AWS collection of MCP examples demonstrating client-server interactions with AWS services. **When to choose:** When you're looking for reference implementations of MCP servers on AWS infrastructure. **When NOT to choose:** When you need a ready-to-use MCP server rather than implementation examples. ### [mcp-stata](https://mcpcatalogs.com/en/server/tmonk-mcp-stata) by tmonk · score 46 · ★62 A Model Context Protocol server for Stata that enables AI agents to execute Stata code, inspect data, and retrieve results. **When to choose:** Choose this if you're an economist or researcher already using Stata who wants to integrate AI assistance directly into your statistical workflow. **When NOT to choose:** Don't choose this if you don't have a licensed copy of Stata 17+, as the server requires a local installation of proprietary statistical software. ### [novyx-mcp](https://mcpcatalogs.com/en/server/novyxlabs-novyx-mcp) by novyxlabs · score 46 · ★30 A comprehensive MCP server providing persistent memory and governance for AI agents with 120+ tools. **When to choose:** Choose Novyx-MCP when you need memory persistence across AI agent sessions with governance features and audit trails, especially when working with multiple agents that need shared context. **When NOT to choose:** Avoid if you need an open-source solution without cloud dependencies, or if you require features not covered by their tiered pricing (e.g., enterprise-only interventions, pro+ threat intelligence). ### [predictive-maintenance-mcp](https://mcpcatalogs.com/en/server/lgdimaggio-predictive-maintenance-mcp) by LGDiMaggio · score 46 · ★38 MCP server enabling AI assistants to analyze vibration data, detect machinery faults, and generate diagnostic reports. **When to choose:** Choose this server for industrial predictive maintenance workflows requiring vibration analysis, fault detection, and diagnostic report generation through natural language interactions. **When NOT to choose:** Not suitable for teams needing real-time monitoring capabilities or integration with non-vibration-based predictive maintenance systems. ### [nia](https://mcpcatalogs.com/en/server/nozomio-labs-nia) by nozomio-labs · score 46 · ★69 Nia is an MCP server providing AI agents with indexed documentation, code, and file search capabilities. **When to choose:** Choose Nia when you need comprehensive search capabilities across multiple knowledge sources for your AI agents, especially if you're already using Cursor, Claude Code, or other supported agents. **When NOT to choose:** Don't choose Nia if you need an open-source solution with no vendor lock-in, or if you require free usage beyond the very limited free tier (only 50 queries per month). ### [blatant-why](https://mcpcatalogs.com/en/server/001tmf-blatant-why) by 001TMF · score 46 · ★66 Multi-agent biologic design platform with 11 MCP servers for protein design campaigns. **When to choose:** Choose Blatant-Why when you need a comprehensive, end-to-end protein design solution that integrates multiple tools and AI models through Claude Code without requiring specialized infrastructure. **When NOT to choose:** Don't choose this if you're looking for a standalone tool that works outside of Claude Code, or if you require more control over the underlying computational methods than the abstraction layer provides. ### [interactive-mcp](https://mcpcatalogs.com/en/server/ttommyth-interactive-mcp) by ttommyth · score 46 · ★345 interactive-mcp is a local, cross-platform MCP server that enables interactive communication between AI agents and users through prompts, notifications, and persistent chat sessions. **When to choose:** Choose this when your AI application needs to interact directly with users through prompts and notifications, especially during development workflows or pair programming scenarios. **When NOT to choose:** Avoid this for server-to-server interactions where user input isn't needed, or when running in environments without direct access to the user's local machine. ### [adaptive-cards-mcp](https://mcpcatalogs.com/en/server/vikrantsingh01-adaptive-cards-mcp) by VikrantSingh01 · score 46 · ★26 AI-powered Adaptive Card MCP server with 9 tools for generating, validating, and optimizing cards for Teams, Outlook, Copilot, and ChatGPT. **When to choose:** Choose this server when working with Microsoft Teams, Outlook, or other Microsoft platforms and need to generate or optimize Adaptive Cards with proper validation and accessibility features. **When NOT to choose:** Avoid this if you're primarily working with non-Microsoft platforms that don't support Adaptive Cards, or if you need more extensive customization beyond Microsoft's ecosystem constraints. ### [ha-mcp-for-xiaozhi](https://mcpcatalogs.com/en/server/c1pher-cn-ha-mcp-for-xiaozhi) by c1pher-cn · score 46 · ★234 Home Assistant MCP server for Xiaozhi AI enabling direct integration with voice assistants. **When to choose:** Choose this if you're already using Home Assistant and Xiaozhi AI and want direct voice control over your smart devices without intermediate proxies. **When NOT to choose:** Not recommended if you need write access to Home Assistant (this appears to be read-only only) or if you're not already using Xiaozhi AI as your primary voice assistant. ### [openfoam-mcp-server](https://mcpcatalogs.com/en/server/webworn-openfoam-mcp-server) by webworn · score 46 · ★97 Intelligent CFD education system using Socratic questioning and OpenFOAM integration with AI-driven problem solving. **When to choose:** Educators and CFD students who need an interactive learning system that uses Socratic questioning and context engineering to teach computational fluid dynamics concepts. **When NOT to choose:** Professional CFD engineers needing a fully integrated production environment, as this server focuses more on education than advanced simulation workflows. ### [cesium-ai-integrations](https://mcpcatalogs.com/en/server/cesiumgs-cesium-ai-integrations) by CesiumGS · score 46 · ★70 Reference integrations connecting Cesium ecosystem with AI systems including MCP tools. **When to choose:** Choose this when you're building geospatial applications that need to integrate AI capabilities with 3D visualization and data processing. **When NOT to choose:** Not ideal for projects outside the geospatial domain or those requiring a different protocol than MCP for AI integration. ### [Agent-Fusion](https://mcpcatalogs.com/en/server/krokozyab-agent-fusion) by krokozyab · score 46 · ★66 Local RAG semantic search engine providing AI agents access to code, documentation and PDFs via MCP protocol. **When to choose:** Choose Agent Fusion when you need a local RAG search engine for AI agents to query your codebase and documentation without internet connectivity or cloud dependencies. **When NOT to choose:** Don't choose Agent Fusion if you need cloud-based search capabilities, real-time collaboration features, or require integration with non-local file systems. ### [join.cloud](https://mcpcatalogs.com/en/server/kushneryk-join-cloud) by kushneryk · score 46 · ★66 A collaboration platform for AI agents to interact in real-time rooms via MCP, A2A protocols. **When to choose:** Choose Join.cloud when you need real-time collaboration between AI agents with persistent message history, file sharing through git, and room management features like roles and permissions. **When NOT to choose:** Avoid Join.cloud if you need a simpler messaging solution without file collaboration or if the AGPL-3.0 license is incompatible with your project's requirements. ### [toonify-mcp](https://mcpcatalogs.com/en/server/pcircle-ai-toonify-mcp) by PCIRCLE-AI · score 46 · ★63 Context compression plugin for Claude Code that trims large JSON, logs, and source files. **When to choose:** Choose this when working with Claude Code and frequently encountering context window issues due to large JSON, logs, or stack traces. **When NOT to choose:** Don't choose this if you need exact original formatting preserved or for very short text content that doesn't benefit from compression. ### [PyWry](https://mcpcatalogs.com/en/server/deeleeramone-pywry) by deeleeramone · score 46 · ★81 PyWry is a cross-platform rendering engine that provides MCP server tools for creating interactive widgets and dashboards. **When to choose:** Choose PyWry when you need to build interactive UI components for AI applications and want to deploy them across desktop, web, and notebook environments with minimal code changes. **When NOT to choose:** Don't choose PyWry if you need complex form controls beyond basic UI elements, or if you require extensive native OS integration that webview cannot provide. ### [thoughtbox](https://mcpcatalogs.com/en/server/kastalien-research-thoughtbox) by Kastalien-Research · score 46 · ★59 Thoughtbox is an MCP server that creates an intention ledger for AI agents to coordinate through shared workspaces with auditable reasoning. **When to choose:** Choose Thoughtbox when you need collaborative reasoning between multiple AI agents with auditable decision trails and visual exploration of complex thought processes. **When NOT to choose:** Avoid if you need a simple single-agent task automation tool, or if you're not comfortable with Docker-based deployment and a more complex multi-agent workflow pattern. ### [optuna-mcp](https://mcpcatalogs.com/en/server/optuna-optuna-mcp) by optuna · score 46 · ★76 Optuna MCP server automates hyperparameter optimization and analysis via interactive tools. **When to choose:** Choose Optuna MCP when you need to integrate automated hyperparameter optimization into an MCP-enabled chat interface, especially for machine learning workflows. **When NOT to choose:** Avoid if you need optimization capabilities outside of MCP clients or require advanced optimization algorithms not supported by Optuna. ### [rlm-claude](https://mcpcatalogs.com/en/server/encreor-rlm-claude) by EncrEor · score 46 · ★35 RLM is an MCP server providing persistent memory for Claude Code with auto-save, semantic search, and multi-project organization. **When to choose:** Choose RLM if you work with Claude Code and need persistent memory across sessions, especially for long-term projects where context continuity is crucial. **When NOT to choose:** Don't choose RLM if you need memory for non-Claude environments, or if you prefer all-in-one solutions like MemGPT that come with their own runtime. ### [agentanycast](https://mcpcatalogs.com/en/server/agentanycast-agentanycast) by AgentAnycast · score 46 · ★76 A P2P runtime that enables AI agents to communicate across networks without public IPs, with an MCP server integration. **When to choose:** Choose AgentAnycast when you need to connect AI agents across networks without public IPs, especially for distributed systems requiring end-to-end encryption and skill-based routing. **When NOT to choose:** Avoid if you need direct control over networking infrastructure or if your use case requires integration with existing systems that don't support the A2A protocol. ### [teslamate-mcp](https://mcpcatalogs.com/en/server/cobanov-teslamate-mcp) by cobanov · score 46 · ★127 TeslaMate MCP server provides AI assistants with secure access to Tesla vehicle data through 18 predefined queries and custom SQL. **When to choose:** Choose this server if you're a Tesla owner using TeslaMate and want to query your vehicle data through AI assistants without writing SQL yourself. **When NOT to choose:** Don't choose this if you need write access to your Tesla data or if you don't already have TeslaMate running with PostgreSQL. ### [AgentRelay](https://mcpcatalogs.com/en/server/mnemox-ai-agentrelay) by mnemox-ai · score 46 · ★63 AgentRelay is an MCP server that coordinates verified microtasks between AI agents to optimize idle compute capacity. **When to choose:** Teams with multiple AI agents wanting to efficiently utilize idle capacity and create a reputation-based system for verified work output. **When NOT to choose:** You need a simple, task-based approach without the complexity of validation rules and reputation tracking. ### [claude_code-gemini-mcp](https://mcpcatalogs.com/en/server/raiansar-claude-code-gemini-mcp) by RaiAnsar · score 46 · ★242 An MCP server bridging Claude Code with Google's Gemini for AI collaboration and code assistance. **When to choose:** Choose this when you want to leverage Gemini's AI capabilities within Claude Code for code reviews and brainstorming without leaving your development environment. **When NOT to choose:** Don't choose this if you need advanced Gemini features not exposed by this MCP server, or if you're concerned about vendor lock-in to Google's Gemini platform. ### [mcp-memory-libsql](https://mcpcatalogs.com/en/server/spences10-mcp-memory-libsql) by spences10 · score 46 · ★84 High-performance persistent memory system for MCP with vector search and knowledge graph capabilities. **When to choose:** Choose this when you need a persistent memory system with vector search capabilities for AI agents, especially if you're already using libSQL or SQLite databases. **When NOT to choose:** Avoid if you need read/write access to existing databases or require more advanced semantic search capabilities beyond basic vector search and relationship management. ### [glif-mcp-server](https://mcpcatalogs.com/en/server/glifxyz-glif-mcp-server) by glifxyz · score 46 · ★122 MCP server enabling AI workflow execution from glif.app within your LLM client. **When to choose:** Choose this server if you need to integrate AI image generation or multimedia workflows into your LLM conversations, especially if you're already using glif.app. **When NOT to choose:** Avoid this server if you don't want vendor lock-in to glif.app, or if you need AI workflows beyond what glif supports. ### [mcp_chatbot](https://mcpcatalogs.com/en/server/keli-wen-mcp-chatbot) by keli-wen · score 46 · ★247 A Python chatbot implementation compatible with MCP, supporting terminal and Streamlit interfaces with customizable LLM integration. **When to choose:** Choose this implementation if you're developing chatbots that need to interact with MCP tools, especially if you need multiple interface options (CLI, terminal, web) and want to integrate with custom LLM providers. **When NOT to choose:** Don't choose this if you need a production-ready MCP server implementation rather than a chatbot framework, or if you require advanced authentication and security features. ### [magg](https://mcpcatalogs.com/en/server/sitbon-magg) by sitbon · score 46 · ★133 MCP Aggregator server that manages multiple MCP servers, enabling dynamic tool discovery and configuration. **When to choose:** Choose Magg when you need to manage multiple MCP servers or want AI systems to dynamically discover and add new capabilities at runtime. **When NOT to choose:** Don't choose Magg if you only need a single MCP server or are looking for a specific functionality rather than server management capabilities. ### [daisyui-mcp](https://mcpcatalogs.com/en/server/birdseyevue-daisyui-mcp) by birdseyevue · score 46 · ★67 A token-efficient MCP server providing DaisyUI component documentation to AI assistants. **When to choose:** Choose this server when building applications with DaisyUI and need efficient access to component documentation through AI assistants. **When NOT to choose:** Don't choose if you need access to DaisyUI's latest documentation instantly, as updates require manual execution of the update script. ### [luma-mcp](https://mcpcatalogs.com/en/server/jochenyang-luma-mcp) by JochenYang · score 46 · ★59 Multi-Model Visual Understanding MCP Server providing image analysis capabilities to AI coding models that don't support native vision. **When to choose:** Choose Luma MCP when you need visual understanding capabilities for AI coding models that don't natively support image processing, especially when working with code screenshots, UI designs, or document images. **When NOT to choose:** Avoid if you need model-specific features not supported by Luma's unified approach, or if you require handling images larger than 10MB with complex processing needs. ### [altk-evolve](https://mcpcatalogs.com/en/server/agenttoolkit-altk-evolve) by AgentToolkit · score 46 · ★84 Evolve provides AI agents with memory and learning capabilities through MCP server integration. **When to choose:** Choose Evolve when you need AI agents that learn and improve from past interactions, particularly for coding or complex multi-step tasks where agents would otherwise repeat the same mistakes. **When NOT to choose:** Avoid Evolve if you need a simple, lightweight solution without persistent memory or if you're not using Python-based agents, as it requires Python 3.12+ and has multiple dependencies. ### [mcp-bridgekit](https://mcpcatalogs.com/en/server/mkbhardwas12-mcp-bridgekit) by mkbhardwas12 · score 46 · ★60 A bridge that converts MCP stdio servers to HTTP endpoints with session pooling and background job handling. **When to choose:** Choose this when building web applications that need to integrate with MCP servers and require session management, timeout handling, and background processing. **When NOT to choose:** Don't choose this for simple MCP integrations that don't require HTTP bridging or if you need a solution without external dependencies like Redis. ### [speclock](https://mcpcatalogs.com/en/server/sgroy10-speclock) by sgroy10 · score 46 · ★24 SpecLock is an AI constraint engine with 51 MCP tools that enforces project rules across AI coding sessions. **When to choose:** Choose SpecLock when you need to enforce constraints across AI coding sessions and protect critical systems from unintended modifications, especially in teams or projects with strict governance requirements. **When NOT to choose:** Avoid SpecLock if you're looking for simple AI memory tools without enforcement capabilities, or if you need very fine-grained control over individual AI actions rather than high-level constraints. ### [remind](https://mcpcatalogs.com/en/server/sandst1-remind) by sandst1 · score 46 · ★78 Remind is an AI agent memory layer that extracts and maintains generalized concepts rather than verbatim text. **When to choose:** Choose Remind when you need an AI memory system that consolidates experiences into abstract knowledge rather than just storing raw text. **When NOT to choose:** Don't choose Remind if you need verbatim text recall, as it stores generalized concepts and may lose original wording. ### [drawio2go](https://mcpcatalogs.com/en/server/menghuan1918-drawio2go) by Menghuan1918 · score 46 · ★199 DrawIO2Go is an AI-powered diagram editor with MCP server integration for canvas content version management. **When to choose:** Choose DrawIO2Go when you need an AI-powered diagramming tool with version control and want to integrate your diagram workflow with other applications through MCP. **When NOT to choose:** Avoid DrawIO2Go if you need enterprise-level multi-user support or require extensive integration options beyond the MCP canvas version management feature. ### [context-engineering](https://mcpcatalogs.com/en/server/timothywarner-org-context-engineering) by timothywarner-org · score 45 · ★23 Production-ready MCP server implementing semantic memory with hybrid RAG and CoALA memory tiers for AI assistants. **When to choose:** Choose this when you need a comprehensive memory system for AI assistants, particularly if you're implementing production solutions with semantic memory, hybrid RAG, and need educational examples of MCP implementation. **When NOT to choose:** Don't choose this if you need a simple MCP server without the complexity of a full memory architecture, or if you're looking for a solution that works with AI models other than Claude. ### [korean-contracts](https://mcpcatalogs.com/en/server/kimlawtech-korean-contracts) by kimlawtech · score 45 · ★42 MCP server for generating Korean legal contracts with privacy protection features. **When to choose:** Choose this system if you need to create Korean contracts with the latest legal requirements and want your personal information protected through an MCP privacy server. **When NOT to choose:** This system is limited to Korean contracts only and doesn't support international contract templates. It also requires Claude Desktop/Codex CLI and doesn't have a standalone web interface. ### [OmniClip-RAG](https://mcpcatalogs.com/en/server/msjsc001-omniclip-rag) by msjsc001 · score 45 · ★38 Local-first RAG desktop app and MCP server for searching private Markdown, PDF and 1290+ document formats. **When to choose:** Choose this when you need a local, privacy-first solution for AI to query your personal documentation without uploading sensitive data to third-party services. **When NOT to choose:** Not suitable if you need cross-platform support, write capabilities to your knowledge base, or require integration with cloud-based document repositories. ### [tokio-prompt-orchestrator](https://mcpcatalogs.com/en/server/mattbusel-tokio-prompt-orchestrator) by Mattbusel · score 45 · ★55 Multi-core Tokio-native orchestration for LLM pipelines with MCP server support. **When to choose:** Choose this when you need a production-ready, multi-core orchestration system for LLM pipelines with advanced features like provider arbitrage and request deduplication. **When NOT to choose:** Avoid this if you need a simple, single-provider solution or if you're not using Rust/Tokio in your tech stack. ### [mcp-playground](https://mcpcatalogs.com/en/server/elkhn-mcp-playground) by Elkhn · score 45 · ★45 A Streamlit-based chat playground with plug-and-play MCP server support for various LLM providers. **When to choose:** Choose this playground when you need to quickly prototype and test MCP tool integrations across multiple LLM providers with minimal setup. **When NOT to choose:** Don't choose this for production deployments as it's designed for development/testing purposes and lacks robust authentication mechanisms. ### [agience-core](https://mcpcatalogs.com/en/server/agience-agience-core) by Agience · score 45 · ★39 Agience is an MCP-native platform that structures AI outputs into durable, versioned artifacts with provenance tracking. **When to choose:** Choose Agience when you need enterprise-grade trust and auditability for AI outputs, with structured artifact tracking and provenance for compliance and collaborative workflows. **When NOT to choose:** Avoid Agience if you need simple file-based storage, prefer permissive licensing over AGPL, or want a lightweight solution without the complexity of its artifact model. ### [bonnard-cli](https://mcpcatalogs.com/en/server/bonnard-data-bonnard-cli) by bonnard-data · score 45 · ★48 Bonnard CLI provides an agent-native analytics platform with MCP server integration for governed data queries. **When to choose:** Teams already using multiple data warehouses who want a consistent semantic layer for AI agents without building a custom data pipeline. **When NOT to choose:** You need write access to your database or require complex data transformations - Bonnard is focused on read-only access and semantic consistency. ### [WAIaaS](https://mcpcatalogs.com/en/server/minhoyoo-iotrust-waiaas) by minhoyoo-iotrust · score 45 · ★26 WAIaaS is a self-hosted wallet daemon that enables AI agents to perform blockchain transactions securely through MCP integration. **When to choose:** Choose WAIaaS when building AI applications that need to interact with blockchain assets but require better security than exposing private keys, with policy controls for transaction approval. **When NOT to choose:** Avoid WAIaaS if you need simple direct key management without policy layers, or if you're building on a blockchain not supported by its adapters (only Solana and EVM chains). ### [Generative-UI-MCP](https://mcpcatalogs.com/en/server/op7418-generative-ui-mcp) by op7418 · score 45 · ★52 MCP server for generating interactive visualizations with design guidelines loaded on demand. **When to choose:** Choose this server when you need AI models to generate interactive visualizations consistently and want to avoid bloating system prompts with design rules. **When NOT to choose:** Don't choose this if you need visualization types beyond the supported modules (charts, diagrams, mockups, art, interactive UI) or require extensive customization of design systems. ### [crash-mcp](https://mcpcatalogs.com/en/server/nikkoxgonzales-crash-mcp) by nikkoxgonzales · score 45 · ★69 CRASH is an MCP server for structured reasoning with step validation, branching, and confidence tracking. **When to choose:** Choose CRASH for complex, multi-step problem-solving tasks requiring systematic breakdown of issues, especially when you need to explore multiple solution paths or track confidence through reasoning steps. **When NOT to choose:** Avoid CRASH for simple, single-step tasks or pure information retrieval where its structured reasoning overhead provides no benefit. ### [mcp-desktop](https://mcpcatalogs.com/en/server/http4k-mcp-desktop) by http4k · score 45 · ★40 A robust Kotlin-based MCP desktop client supporting multiple transport protocols and authentication methods for connecting to MCP servers. **When to choose:** Choose this when you need a robust desktop client connecting AI applications to MCP servers with multiple protocols and authentication options, especially for Cursor integration. **When NOT to choose:** Avoid if you need a commercial license for commercial use, as it uses the http4k Commercial License which requires payment for commercial applications. ### [local-skills-mcp](https://mcpcatalogs.com/en/server/kdpa-llc-local-skills-mcp) by kdpa-llc · score 45 · ★28 Universal MCP server that lets any AI agent access expert skills from your local filesystem with lazy loading. **When to choose:** Choose this when you need to create and maintain expert skills that can be used across multiple AI systems and want efficient context usage through lazy loading. **When NOT to choose:** Avoid this if you need centralized skill management across a team or require more sophisticated skill versioning and collaboration features. ### [open-streetmap-mcp](https://mcpcatalogs.com/en/server/jagan-shanmugam-open-streetmap-mcp) by jagan-shanmugam · score 45 · ★191 A comprehensive OpenStreetMap MCP server enabling geocoding, routing, POI searches, and neighborhood analysis through location-based tools. **When to choose:** Choose this when you need location-based data and services in your LLM applications, particularly for routing, POI discovery, and geospatial analysis. **When NOT to choose:** Don't choose this if you need real-time traffic data, advanced routing algorithms, or location data from sources other than OpenStreetMap. ### [ctxvault](https://mcpcatalogs.com/en/server/filippo-venturini-ctxvault) by Filippo-Venturini · score 45 · ★56 CtxVault provides typed memory infrastructure for AI agents with semantic/procedural vault isolation. **When to choose:** Choose CtxVault when you need structured isolation between multiple AI agents, persistent memory across sessions, and full human control over agent knowledge bases. **When NOT to choose:** Don't choose CtxVault if you need a simple in-memory context window, require cloud-based storage, or need extensive authentication and security features beyond basic access control. ### [MCPollinations](https://mcpcatalogs.com/en/server/pinkpixel-dev-mcpollinations) by pinkpixel-dev · score 45 · ★41 MCP server for AI assistants to generate images, text, and audio through Pollinations APIs. **When to choose:** Choose MCPollinations when you need multimodal content generation capabilities in your AI workflow without complex setup or API management. **When NOT to choose:** Avoid if you need proprietary model access or require strict content filtering, as MCPollinations relies on the external Pollinations API with limited control over content policies. ### [subcog](https://mcpcatalogs.com/en/server/zircote-subcog) by zircote · score 45 · ★22 Persistent memory system for AI coding assistants with MCP integration, hybrid search, and knowledge graph features. **When to choose:** Choose Subcog when you need a persistent memory system specifically for coding assistants, particularly if you're already using Rust or Claude Code and want seamless IDE integration. **When NOT to choose:** Avoid Subcog if you need a general-purpose knowledge management system, or if you require authentication/authorization beyond simple JWT tokens for the HTTP transport mode. ### [eShopLite](https://mcpcatalogs.com/en/server/azure-samples-eshoplite) by Azure-Samples · score 45 · ★162 eShopLite is an .NET e-commerce reference app with integrated MCP server for AI interactions. **When to choose:** Choose eShopLite when building .NET-based eCommerce applications that need to integrate AI capabilities with MCP server architecture. **When NOT to choose:** Don't choose eShopLite for non-.NET projects, simple storefronts without AI features, or if you need production-ready MCP implementations with advanced security features. ### [deepseek-mcp](https://mcpcatalogs.com/en/server/arizen-dev-deepseek-mcp) by arizen-dev · score 45 · ★48 A tiny MCP server exposing DeepSeek as a cheap supervised worker for Claude Code, Codex, or any MCP client. **When to choose:** Choose this server when you need a cost-effective AI worker for mechanical tasks, classification, or data transformation, especially if you already have DeepSeek API access. **When NOT to choose:** Don't choose this for complex architecture design, security policy decisions, or when you need a final polished output without human review. ### [memcp](https://mcpcatalogs.com/en/server/maydali28-memcp) by maydali28 · score 45 · ★18 A persistent memory MCP server for Claude Code that implements the Recursive Language Model framework to store knowledge across sessions. **When to choose:** Choose MemCP when you're working on long-term projects with Claude Code where maintaining context across compactions and sessions is crucial, especially with large codebases or multiple documents. **When NOT to choose:** Avoid MemCP if you need cross-platform support beyond Python, prefer cloud-based solutions, or need access to your memory data through web interfaces or APIs. ### [datawrapper-mcp](https://mcpcatalogs.com/en/server/palewire-datawrapper-mcp) by palewire · score 45 · ★38 An MCP server for creating Datawrapper charts through AI assistants, with comprehensive chart management tools. **When to choose:** Choose this server when you need AI-powered chart creation with Datawrapper and want to leverage AI assistants for data visualization without switching between tools. **When NOT to choose:** Don't choose this if you need to work with chart creation tools other than Datawrapper, or if you require features not supported by Datawrapper's API. ### [grok-faf-mcp](https://mcpcatalogs.com/en/server/wolfe-jam-grok-faf-mcp) by Wolfe-Jam · score 45 · ★16 Persistent project context for xAI Grok using IANA-registered .faf format with URL-based MCP server. **When to choose:** Choose this server if you're working with xAI Grok and need persistent project context without repetitive explanations, especially for teams maintaining consistent understanding across multiple sessions. **When NOT to choose:** Don't choose this if you need write access to your project files (this is primarily a read-context server) or if you're working with AI platforms other than Grok/xAI. ### [arifos](https://mcpcatalogs.com/en/server/ariffazil-arifos) by ariffazil · score 45 · ★43 Constitutional AI governance kernel with 13 MCP tools for governed agentic execution. **When to choose:** Choose arifOS when building secure, audited AI systems requiring constitutional guardrails and transparent verification of agent actions. **When NOT to choose:** Not suitable for simple, unregulated AI applications or environments where the overhead of constitutional validation isn't justified. ### [VSCode-Perplexity-MCP](https://mcpcatalogs.com/en/server/automations-project-vscode-perplexity-mcp) by Automations-Project · score 45 · ★50 Perplexity AI search and reasoning integrated as MCP tools for VS Code and 15+ other IDEs through a browser automation approach. **When to choose:** Choose this if you want to integrate Perplexity's AI capabilities directly into your development workflow without needing API keys or leaving your IDE. **When NOT to choose:** Don't choose this if you need programmatic access to AI functions beyond what's available in Perplexity's interface, or if you're concerned about browser automation overhead. ### [congressMCP](https://mcpcatalogs.com/en/server/amurshak-congressmcp) by amurshak · score 45 · ★30 MCP server providing 91+ tools for accessing live U.S. Congressional data through Congress.gov API. **When to choose:** Choose CongressMCP when you need access to current U.S. legislative data through an MCP interface, particularly for research, journalism, or applications requiring congressional information. **When NOT to choose:** Avoid CongressMCP if you need non-U.S. legislative data, require write access to congressional records, or need data before 1989 (Congress.gov coverage limit). ### [mcp-dashboards](https://mcpcatalogs.com/en/server/kyurish-mcp-dashboards) by KyuRish · score 45 · ★22 MCP server that renders interactive data visualizations and dashboards inside AI conversations. **When to choose:** Choose MCP Dashboards when you need rich data visualization within AI conversations without switching context to a separate BI tool. **When NOT to choose:** Avoid if you need a BI platform with advanced data transformation, write capabilities to databases, or require commercial licensing for a competitive product. ### [chatgpt-app-typescript-template](https://mcpcatalogs.com/en/server/pomerium-chatgpt-app-typescript-template) by pomerium · score 45 · ★16 Production-ready MCP app template for building ChatGPT-style tools with React UI widgets. **When to choose:** Choose this template when starting a new MCP application that needs a rich UI with React components and a well-structured production-ready codebase. **When NOT to choose:** Avoid if you need a simple tool without UI components or if you're not using TypeScript and React in your stack. ### [octave-mcp](https://mcpcatalogs.com/en/server/elevanaltd-octave-mcp) by elevanaltd · score 45 · ★51 OCTAVE MCP server provides structured document infrastructure with canonicalization, schema validation, and mythological compression for AI artifacts. **When to choose:** Choose OCTAVE when working with AI documents that need to maintain consistency across multiple agents or compression steps, especially when token efficiency and deterministic output are critical. **When NOT to choose:** Don't choose OCTAVE for single-step prompts, freeform prose documents, or code output where structured formatting isn't needed. ### [fortuneteller](https://mcpcatalogs.com/en/server/hjsh200219-fortuneteller) by hjsh200219 · score 44 · ★34 An MCP server for Korean traditional Saju fortune-telling with comprehensive analysis tools. **When to choose:** Choose this MCP server if you're interested in Korean traditional fortune-telling or need detailed Saju analysis for personal guidance or relationship compatibility. **When NOT to choose:** Don't choose this if you're looking for Western astrology or other cultural fortune-telling systems, as it's specifically focused on Korean Saju traditions. ### [hass-mcp-server](https://mcpcatalogs.com/en/server/ganhammar-hass-mcp-server) by ganhammar · score 44 · ★34 Home Assistant MCP server with HTTP transport for remote access via OAuth or long-lived tokens. **When to choose:** Choose this if you want remote access to Home Assistant via Claude.ai or any web-based AI assistant, especially when OAuth authentication is needed. **When NOT to choose:** Don't choose this if you only need local access to Home Assistant with SSE transport or if you prefer the official Home Assistant MCP integration. ### [metmuseum-mcp](https://mcpcatalogs.com/en/server/mikechao-metmuseum-mcp) by mikechao · score 44 · ★26 An MCP server providing access to The Metropolitan Museum of Art's collection with search, retrieval and visualization tools. **When to choose:** Choose this server when you need access to the Metropolitan Museum of Art's collection for AI-powered art exploration and research. **When NOT to choose:** Don't choose this if you need access to museums other than The Met, or if you require features beyond search and retrieval of artwork information. ### [hwp-mcp](https://mcpcatalogs.com/en/server/treesoop-hwp-mcp) by treesoop · score 44 · ★26 MCP server enabling AI agents to read, edit, and create Korean HWP/HWPX documents with 34 specialized tools. **When to choose:** Choose this MCP server when working with Korean Hangul documents and need AI agents to read, analyze, or modify them, especially if you're already using Claude/Cursor/ChatGPT. **When NOT to choose:** Don't choose this if you primarily need to work with .hwp files (not .hwpx) as write support is limited, or if you need features not yet implemented like chart handling or advanced style definitions. ### [deep-research-mcp](https://mcpcatalogs.com/en/server/pinkpixel-dev-deep-research-mcp) by pinkpixel-dev · score 44 · ★27 A research-focused MCP server using Tavily APIs to gather and structure web information for LLM documentation generation. **When to choose:** Choose this server when you need comprehensive web research capabilities with structured output for LLM document generation, especially if you're already using Tavily's APIs. **When NOT to choose:** Avoid this server if you need offline research capabilities or if you prefer not to rely on Tavily's services for your research needs. ### [openalex-research-mcp](https://mcpcatalogs.com/en/server/oksure-openalex-research-mcp) by oksure · score 44 · ★26 MCP server providing access to 240M+ scholarly works through 31 specialized tools for research analysis and literature reviews. **When to choose:** Choose this server when you need to perform academic research analysis with credible sources through an MCP interface, particularly for literature reviews, citation analysis, and trend tracking. **When NOT to choose:** Don't choose this server if you need write capabilities to academic databases (this is read-only only) or if you're working outside the academic research domain. ### [needle-mcp](https://mcpcatalogs.com/en/server/needle-ai-needle-mcp) by needle-ai · score 44 · ★100 Needle MCP Server enables RAG functionality with document management and semantic search for LLMs. **When to choose:** Choose this when you need semantic search capabilities across your document collections and are already using or willing to use the Needle platform. **When NOT to choose:** Avoid if you need self-hosted solutions without vendor lock-in, as this requires integration with Needle's proprietary service. ### [civic-ai-tools](https://mcpcatalogs.com/en/server/npstorey-civic-ai-tools) by npstorey · score 44 · ★30 MCP server connecting AI assistants to government open data with anti-hallucination framework. **When to choose:** Choose this when you need to analyze civic/government data using AI assistants and want direct access to datasets without programming expertise. **When NOT to choose:** Don't choose this if you need data sources beyond NYC/Google Data Commons, or if you require higher rate limits for commercial usage. ### [gemini-webapi-mcp](https://mcpcatalogs.com/en/server/andyshaman-gemini-webapi-mcp) by AndyShaman · score 44 · ★27 MCP server for Google Gemini that provides image generation, editing, chat, and file analysis via browser cookies. **When to choose:** Choose this when you want to access Google Gemini's capabilities without dealing with API keys or authentication hassles, especially for image generation with automatic upscaling and watermark removal. **When NOT to choose:** Don't choose this if you need a stable API for production applications, as it relies on reverse-engineering Google's web interface which could break without notice. ### [agentgate](https://mcpcatalogs.com/en/server/agentkitai-agentgate) by agentkitai · score 44 · ★27 Human-in-the-loop approval system for AI agents with MCP integration and policy-based decision routing. **When to choose:** Choose AgentGate when you need human oversight for AI agents that perform sensitive actions and require proper approval workflows with audit trails. **When NOT to choose:** Don't choose AgentGate if you need a lightweight solution without the overhead of a full approval system, or if you're already invested in another workflow platform. ### [lotus-wisdom-mcp](https://mcpcatalogs.com/en/server/linxule-lotus-wisdom-mcp) by linxule · score 44 · ★27 An MCP server implementing the Lotus Sutra wisdom framework for contemplative problem-solving with interactive visualization. **When to choose:** Choose this server for contemplative problem-solving that benefits from a structured, multi-faceted approach inspired by eastern wisdom traditions, especially when you need both analytical thinking and intuitive insights. **When NOT to choose:** Don't choose this if you need quick, straightforward solutions without the contemplative process, or if you prefer a more direct, linear approach to problem-solving. ### [mindbridge-mcp](https://mcpcatalogs.com/en/server/pinkpixel-dev-mindbridge-mcp) by pinkpixel-dev · score 44 · ★34 MindBridge is an AI orchestration MCP server that unifies access to multiple LLM providers through a single API. **When to choose:** Choose MindBridge when you need to orchestrate multiple LLM providers or compare model outputs for critical tasks, especially if you're already using multiple AI services and want a unified interface. **When NOT to choose:** Avoid if you need maximum security control over your LLM interactions or are looking for a lightweight solution with minimal dependencies. ### [local_faiss_mcp](https://mcpcatalogs.com/en/server/nonatofabio-local-faiss-mcp) by nonatofabio · score 44 · ★30 A local FAISS vector store MCP server providing document ingestion and semantic search functionality. **When to choose:** Choose this when you need a local vector store for RAG applications without external dependencies, with support for multiple document formats and optional re-ranking. **When NOT to choose:** Not suitable for very large document collections (>1M documents) or when you need distributed vector storage with high availability. ### [langchain-mcp-tools-py](https://mcpcatalogs.com/en/server/hideya-langchain-mcp-tools-py) by hideya · score 44 · ★28 Python utility converting MCP server tools into LangChain-compatible tools with support for local and remote servers. **When to choose:** Choose this when you need a simple way to integrate MCP servers with LangChain and only require basic tool functionality. **When NOT to choose:** Don't choose this if you need full MCP protocol support with Resources, Prompts, or multi-content type results (images, etc.). ### [IntelliConnect](https://mcpcatalogs.com/en/server/ruanrongman-intelliconnect) by ruanrongman · score 44 · ★118 An AI-powered IoT platform with MCP protocol support for smart device management and agent capabilities. **When to choose:** Choose IntelliConnect if you're building AI-powered IoT applications that require integration with multiple devices, voice capabilities, and need to extend functionality through the MCP protocol. **When NOT to choose:** Avoid this platform if you need a simple MQTT broker without AI capabilities, or if you prefer a pure Python-based solution as this is primarily Java-based. ### [MiniMax-MCP-JS](https://mcpcatalogs.com/en/server/minimax-ai-minimax-mcp-js) by MiniMax-AI · score 44 · ★122 JavaScript/TypeScript MCP server providing MiniMax AI capabilities including image/video generation, TTS, and voice cloning. **When to choose:** Choose this MCP server when you need integrated access to MiniMax's AI capabilities including text-to-speech, image generation, and video creation through the MCP protocol. **When NOT to choose:** Don't choose this if you need to work with AI providers other than MiniMax, as it's a vendor-locked solution. ### [mmx-mcp-server](https://mcpcatalogs.com/en/server/zth0828-mmx-mcp-server) by zth0828 · score 44 · ★31 MCP server providing MiniMax's multi-modal capabilities including text, search, image processing, voice synthesis and more. **When to choose:** Choose this MCP server if you need to integrate MiniMax's multi-modal AI capabilities (text, images, audio, video) into your MCP-enabled AI workflows, especially if you're already using MiniMax's services. **When NOT to choose:** Avoid this if you need access to MiniMax models outside of MCP, require features not available in mmx CLI, or need to minimize vendor lock-in to MiniMax's ecosystem. ### [DINO-X-MCP](https://mcpcatalogs.com/en/server/idea-research-dino-x-mcp) by IDEA-Research · score 44 · ★114 DINO-X MCP server enables LLMs with visual perception through image object detection, localization, and captioning APIs. **When to choose:** Choose DINO-X MCP when you need fine-grained visual understanding with structured outputs for multimodal AI applications, especially when working with object detection, localization, and image captioning tasks. **When NOT to choose:** Avoid if you need local-only processing without API dependency, or if you require processing image files via local paths in HTTP mode. ### [extract-llms-docs](https://mcpcatalogs.com/en/server/nirholas-extract-llms-docs) by nirholas · score 44 · ★30 An MCP server that extracts llms.txt documentation from websites for AI agents. **When to choose:** Choose this MCP server when you need to extract and organize documentation from websites supporting llms.txt standard for AI assistants and RAG systems. **When NOT to choose:** Don't choose this if you need to extract documentation from websites that don't support the llms.txt standard, as it relies on this specific format for proper parsing. ### [mcp-server-chatsum](https://mcpcatalogs.com/en/server/chatmcp-mcp-server-chatsum) by chatmcp · score 44 · ★1,033 MCP server for querying and summarizing chat messages with database integration. **When to choose:** Choose this server if you need to query and summarize chat message history and have your chats stored in a compatible database. **When NOT to choose:** Don't choose this if you don't have your chat messages in a supported database format or if you need integration with messaging platforms other than what's supported. ### [mcp-server](https://mcpcatalogs.com/en/server/dollhousemcp-mcp-server) by DollhouseMCP · score 44 · ★33 A free, open-source MCP server for dynamic custom persona management with community-shared templates and skills. **When to choose:** Choose DollhouseMCP when you need extensive customization of AI behavior with modular personas, skills, and agents, especially if you're already using an MCP-compatible client like Claude Code or Claude Desktop. **When NOT to choose:** Avoid DollhouseMCP if you need simple, straightforward customization without the complexity of multiple element types or if you're locked into non-MCP AI platforms that don't support external tool integrations. ### [mcp-server-llamacloud](https://mcpcatalogs.com/en/server/run-llama-mcp-server-llamacloud) by run-llama · score 44 · ★86 MCP server connecting to LlamaCloud managed indexes, providing search tools for various document repositories. **When to choose:** Choose this MCP server when you need to query specific document collections stored in LlamaCloud and want to integrate those queries as tools in AI assistants. **When NOT to choose:** Avoid this server if you need direct database access, have concerns about vendor lock-in to LlamaCloud, or prefer self-hosted document indexing solutions. ### [nanobanana-mcp](https://mcpcatalogs.com/en/server/ycse-nanobanana-mcp) by YCSE · score 44 · ★28 An MCP server integrating Gemini's image generation and editing capabilities with Claude Desktop and Claude Code. **When to choose:** Choose this MCP when you need integrated image generation/editing directly within Claude applications and prefer Google's Gemini models over alternatives. **When NOT to choose:** Avoid if you need non-Google model options or require commercial usage rights for generated images, as terms depend on Google's Gemini API terms. ### [atxp](https://mcpcatalogs.com/en/server/atxp-dev-atxp) by atxp-dev · score 44 · ★29 ATXP provides a payment infrastructure and MCP tools for AI agents, including email, SMS, media generation, and code execution. **When to choose:** Choose ATXP when your AI agent needs financial capabilities and wants to pay for external tools like web search, media generation, and communication services without managing API keys. **When NOT to choose:** Don't choose ATXP if you need full control over API keys, require a free tier for testing, or are concerned about vendor lock-in with its proprietary payment system. ### [Memory-Plus](https://mcpcatalogs.com/en/server/yuchen20-memory-plus) by Yuchen20 · score 43 · ★58 Lightweight local RAG memory store for MCP agents with record, retrieve, update, and visualization features. **When to choose:** Choose Memory-Plus when you need a local, privacy-focused memory solution for your AI agents that works across multiple sessions and IDEs without requiring external services. **When NOT to choose:** Don't choose Memory-Plus if you need cloud-based memory sharing across devices, require advanced search capabilities beyond basic keyword matching, or want a solution without external API dependencies like Google's. ### [its-over-9k](https://mcpcatalogs.com/en/server/bumblebiber-its-over-9k) by Bumblebiber · score 43 · ★17 Persistent memory and agent lifecycle framework for Claude Code with MCP tools. **When to choose:** Choose this if you need persistent memory across AI sessions, work on multiple projects, or switch between devices and want AI to maintain context without re-explaining the entire project. **When NOT to choose:** Don't choose this if you need lightweight memory without complex setup, have very simple projects, or prefer solutions with zero configuration requirements. ### [open-brain-template](https://mcpcatalogs.com/en/server/wefilmshit-open-brain-template) by wefilmshit · score 43 · ★23 Persistent memory system for AI agents with visual dashboard and MCP integration. **When to choose:** Choose this if you need structured, persistent memory for AI agents with priority-based loading and are already using or willing to set up Supabase. **When NOT to choose:** Don't choose this if you need a self-contained solution without database dependencies or if you need real-time memory updates with sub-second latency. ### [blender-ai-mcp](https://mcpcatalogs.com/en/server/patrykiti-blender-ai-mcp) by PatrykIti · score 43 · ★24 Production MCP server for Blender with goal-first routing, curated tools, and vision-assisted 3D modeling workflows. **When to choose:** Choose this when you need production-ready AI control over Blender with deterministic verification and structured 3D modeling workflows. **When NOT to choose:** Avoid if you need direct low-level Blender API access or if your workflow requires raw Python script generation rather than structured tool calling. ### [scopus-mcp](https://mcpcatalogs.com/en/server/qwe4559999-scopus-mcp) by qwe4559999 · score 43 · ★25 An MCP server connecting AI assistants to Elsevier Scopus for academic paper search and analysis. **When to choose:** Choose this server if you're working in academic research and need AI assistants to access Scopus database content through MCP. **When NOT to choose:** Don't choose this if you don't have access to an institutional Scopus subscription or if you need search capabilities beyond what Scopus provides. ### [Audrey](https://mcpcatalogs.com/en/server/evilander-audrey) by Evilander · score 43 · ★23 Persistent memory engine for AI agents with MCP integration and REST API. **When to choose:** Choose Audrey when you need persistent memory for AI agents working locally without requiring cloud storage, especially when preventing repeated mistakes from agents is a priority. **When NOT to choose:** Don't choose Audrey if you need a multi-tenant solution with built-in authentication and access control, as it requires manual isolation through different data directories. ### [mcp-openmemory](https://mcpcatalogs.com/en/server/baryhuang-mcp-openmemory) by baryhuang · score 43 · ★72 MCP server enabling Claude to remember conversations and learn over time with SQLite storage. **When to choose:** Choose this when you need personal conversation memory without external dependencies or complex infrastructure setup. **When NOT to choose:** Not suitable for team deployments or sensitive data as it lacks authentication and access controls. ### [turn-mcp](https://mcpcatalogs.com/en/server/shiahonb777-turn-mcp) by shiahonb777 · score 43 · ★24 MCP server that enables infinite conversation turns in a single API request with human-in-the-loop capabilities. **When to choose:** Choose turn-mcp when you need to maintain context across human intervention points in AI workflows, particularly for approval processes or branching decisions where restarting the conversation would lose state. **When NOT to choose:** Avoid this server if you need high-volume human-in-the-loop interactions (limited by max concurrent waits per session) or require a fully managed cloud solution without self-hosting. ### [ClueoMCP](https://mcpcatalogs.com/en/server/clueofoundation-clueomcp) by ClueoFoundation · score 43 · ★74 Universal AI personality layer using Big Five psychology model for consistent behavior across MCP-compatible platforms. **When to choose:** Choose ClueoMCP when you need consistent personality traits across multiple AI platforms and want scientific precision in personality control. **When NOT to choose:** Avoid ClueoMCP if you need open-source implementation without API dependencies or prefer free personality modulation without API key requirements. ### [AskUserQuestionPlus](https://mcpcatalogs.com/en/server/jojojotarou-askuserquestionplus) by JoJoJotarou · score 43 · ★37 MCP server providing a web interface for user questionnaires with keyboard shortcuts and multi-option support. **When to choose:** Choose AskUserQuestionPlus when you need to collect user input interactively within your AI workflow, especially if you value a polished UI with keyboard support and real-time communication. **When NOT to choose:** Don't choose this if you need to integrate with non-HTTP MCP clients, require user authentication, or need to handle more than 20 questions in a single session. ### [AlphaFold-MCP-Server](https://mcpcatalogs.com/en/server/augmented-nature-alphafold-mcp-server) by Augmented-Nature · score 43 · ★34 A comprehensive MCP server providing access to AlphaFold Protein Structure Database with tools for structure retrieval and analysis. **When to choose:** Choose this MCP server when you need programmatic access to AlphaFold protein structure data for research purposes, especially for batch processing or comparative analyses. **When NOT to choose:** Don't choose this if you need offline access to protein structures, as it only retrieves data directly from the AlphaFold API without caching. ### [turn-based-game-mcp](https://mcpcatalogs.com/en/server/github-samples-turn-based-game-mcp) by github-samples · score 43 · ★31 A turn-based games web app with MCP-powered AI opponents featuring Tic-Tac-Toe and Rock Paper Scissors. **When to choose:** Developers looking for a practical MCP example that demonstrates AI interaction with a real application rather than abstract tools. **When NOT to choose:** Production game development needs, as this is explicitly a demo/learning tool, not a scalable gaming platform. ### [bikescout](https://mcpcatalogs.com/en/server/hifly81-bikescout) by hifly81 · score 43 · ★19 AI-powered cycling intelligence tool that predicts terrain conditions and provides tactical route planning through MCP. **When to choose:** Choose BikeScout if you're a cyclist looking for AI-powered route planning with terrain analysis and setup recommendations tailored to your specific bike type and riding conditions. **When NOT to choose:** Avoid this if you need a different activity sport intelligence solution or if you can't comply with the AGPLv3 license for commercial services. ### [robotmem](https://mcpcatalogs.com/en/server/robotmem-robotmem) by robotmem · score 43 · ★24 Robot Memory - Persistent memory system for AI robots with MCP server, hybrid search, and spatial retrieval capabilities. **When to choose:** Choose robotmem for robotic AI systems that need to learn from physical experiences and require structured memory retrieval with spatial and parameter context. **When NOT to choose:** Don't choose this if you need cloud-based memory storage, text-only AI systems without physical parameters, or require GPU acceleration. ### [agenite](https://mcpcatalogs.com/en/server/subeshb1-agenite) by subeshb1 · score 43 · ★69 Agenite is a TypeScript framework for building AI agents with MCP support for connecting to data sources. **When to choose:** Choose Agenite for TypeScript projects requiring type-safe tool integration and multi-agent systems with MCP capabilities. **When NOT to choose:** Don't choose if you need a pure Python solution or require extensive community plugins beyond the current MCP offering. ### [ContextOS](https://mcpcatalogs.com/en/server/itallstartedwithaidea-contextos) by itallstartedwithaidea · score 43 · ★22 A unified MCP context intelligence platform that combines seven foundational repos into a single pip-installable CLI with advanced reasoning capabilities. **When to choose:** Choose ContextOS when building complex AI agents requiring advanced reasoning capabilities beyond simple RAG patterns, especially when dealing with rapidly changing data across multiple sources with different refresh rates. **When NOT to choose:** Avoid if you need a lightweight solution for simple retrieval tasks without the overhead of the cognitive layer, or if your use case doesn't involve data with varying refresh rates that need sophisticated routing strategies. ### [balldontlie-mcp](https://mcpcatalogs.com/en/server/mikechao-balldontlie-mcp) by mikechao · score 43 · ★23 MCP server providing NBA, NFL, and MLB player, team, and game data via Balldontlie API. **When to choose:** Choose this server when you need quick access to NBA, NFL, and MLB sports data through an MCP interface without building custom API integrations. **When NOT to choose:** Don't choose this if you need data from international leagues beyond the three supported American sports, or if you require historical data beyond what Balldontlie API provides. ### [AIRA-SemanticScholar](https://mcpcatalogs.com/en/server/hamid-vakilzadeh-aira-semanticscholar) by hamid-vakilzadeh · score 43 · ★22 MCP server providing AI models with access to Semantic Scholar Academic Graph API for literature search and analysis. **When to choose:** Choose this MCP server if you need academic literature search and citation analysis integrated into AI workflows, especially with Semantic Scholar's comprehensive academic graph data. **When NOT to choose:** Don't choose this server if you need access to other academic databases beyond Semantic Scholar, as it only provides data from the Semantic Scholar API. ### [nexo](https://mcpcatalogs.com/en/server/wazionapps-nexo) by wazionapps · score 43 · ★20 NEXO Brain is a persistent memory system for AI agents that extends their capabilities with semantic RAG and 150+ MCP tools. **When to choose:** Choose NEXO Brain when you need a persistent memory system for AI agents with local-only operation and comprehensive tool integration. **When NOT to choose:** Avoid NEXO Brain if you need cloud-based AI services, real-time collaboration features, or have strict storage limitations on your local device. ### [mcp-agent](https://mcpcatalogs.com/en/server/joshuaalpuerto-mcp-agent) by joshuaalpuerto · score 43 · ★18 TypeScript framework for building AI agents with MCP tools and workflows. **When to choose:** Choose this if you're working in a TypeScript/JavaScript environment and need to build AI agents that can orchestrate multiple MCP tools and implement agent workflows. **When NOT to choose:** Avoid this if you need a more mature ecosystem with broader MCP server support or if you're working primarily in a Python environment where the original mcp-agent would be more suitable. ### [ai-testing-mcp](https://mcpcatalogs.com/en/server/groovy-web-ai-testing-mcp) by groovy-web · score 43 · ★18 Comprehensive MCP server for AI testing, evaluation, and quality assurance with standardized testing methodologies. **When to choose:** Choose AI Testing MCP when you need standardized evaluation of AI models across multiple dimensions and want to integrate testing directly into your AI development workflow. **When NOT to choose:** Avoid if you need testing for custom model architectures not supported by OpenAI/Anthropic, or if you require testing frameworks with extensive visualization capabilities. ### [mcp-gemini-google-search](https://mcpcatalogs.com/en/server/yukukotani-mcp-gemini-google-search) by yukukotani · score 43 · ★79 MCP server providing Google Search functionality using Gemini's built-in Grounding with Google Search feature. **When to choose:** Choose this when you need Google search capabilities within MCP-compliant AI applications, especially if you're already using Google AI or Vertex AI. **When NOT to choose:** Don't choose this if you need search capabilities from other search engines (like Bing or DuckDuckGo) or if you have concerns about vendor lock-in with Google services. ### [bgpt-mcp](https://mcpcatalogs.com/en/server/connerlambden-bgpt-mcp) by connerlambden · score 43 · ★20 A remote MCP server for searching scientific papers with full-text experimental data extraction. **When to choose:** Choose BGPT when you need to search scientific literature with access to raw experimental data rather than just titles and abstracts, and you're working with an MCP-compatible AI tool like Claude or Cursor. **When NOT to choose:** Don't choose BGPT if you need access to papers beyond what's in their database, require offline access, or need a more comprehensive metadata search than what's provided. ### [Mobus](https://mcpcatalogs.com/en/server/mobus-ai-mobus) by mobus-ai · score 43 · ★24 MCP server for searching, previewing, and analyzing datasets across 20+ platforms. **When to choose:** Choose Mobus when you need to discover and analyze datasets across multiple platforms without maintaining separate connections to each one. **When NOT to choose:** Don't choose Mobus if you need write access to datasets or require real-time data updates, as it's read-only with limited API support for some platforms. ### [ab100](https://mcpcatalogs.com/en/server/timothywarner-org-ab100) by timothywarner-org · score 43 · ★20 Certification prep course with MCP-powered Foundry agent implementation and GitHub Copilot study assistant. **When to choose:** Choose this MCP server if you're preparing for Microsoft AB-100 certification or want to learn Agentic AI solutions across the Microsoft ecosystem with integrated study tools. **When NOT to choose:** Don't choose this if you're not interested in Microsoft technologies or don't have access to GitHub Copilot Chat, which is required for the AI assistant functionality. ### [AwesomeSalesforceSkills](https://mcpcatalogs.com/en/server/pranavnagrecha-awesomesalesforceskills) by PranavNagrecha · score 43 · ★20 Comprehensive Salesforce knowledge layer with 38-tool MCP server providing live-org metadata, skills, and agents for AI assistants. **When to choose:** Choose SfSkills when working with Salesforce development or administration and need source-grounded guidance integrated directly into your AI coding assistant. **When NOT to choose:** Don't choose SfSkills if you're not working with Salesforce, as it's specifically tailored for the Salesforce ecosystem with no general-purpose functionality. ### [lightdash_mcp](https://mcpcatalogs.com/en/server/poddubnyoleg-lightdash-mcp) by poddubnyoleg · score 43 · ★19 MCP server connecting AI assistants to Lightdash analytics platform for data discovery and visualization management. **When to choose:** Choose this server if you're already using Lightdash for analytics and want AI assistants to programmatically explore your data, create visualizations, and manage dashboards without leaving your conversational interface. **When NOT to choose:** Avoid this server if you're not using Lightdash or need to connect to other analytics platforms, as it's specifically designed for Lightdash integration only. ### [excalimate](https://mcpcatalogs.com/en/server/excalimate-excalimate) by excalimate · score 43 · ★19 MCP server that turns hand-drawn Excalidraw diagrams into animated videos with AI-driven creation tools. **When to choose:** Choose Excalimate when you need to create animated diagrams with AI assistance, especially if you're already familiar with Excalidraw and want to transform static drawings into engaging animations. **When NOT to choose:** Avoid if you need a highly stable, production-ready tool with long-term support, as the README mentions the code was built with AI and cleanup is ongoing. ### [docmole](https://mcpcatalogs.com/en/server/vigtu-docmole) by Vigtu · score 43 · ★18 An MCP server that lets you query any documentation site with AI, supporting both local RAG and Mintlify-powered sites. **When to choose:** Choose Docmole when you need AI assistants to query documentation sites, especially if you're already using Bun or prefer TypeScript-based solutions. **When NOT to choose:** Don't choose Docmole if you need a solution that doesn't require Bun runtime or if you prefer Python-based RAG systems with more flexibility. ### [statelessagent](https://mcpcatalogs.com/en/server/sgx-labs-statelessagent) by sgx-labs · score 43 · ★17 SAME provides persistent memory for AI coding agents through MCP, enabling context retention across sessions. **When to choose:** Choose SAME when you need persistent memory across AI coding sessions for solo developers or small teams, especially when using multiple AI tools like Claude Code, Cursor, or Windsurf and want to maintain context without cloud dependencies. **When NOT to choose:** Don't choose SAME if you need cloud-based collaboration features, enterprise-grade authentication/SSO, or if you prefer a web-based knowledge management solution rather than local-first approach. ### [deep-research-mcp-server](https://mcpcatalogs.com/en/server/ssdeanx-deep-research-mcp-server) by ssdeanx · score 43 · ★70 MCP server for deep research using Google Gemini 2.5 Flash, providing iterative research capabilities with web grounding. **When to choose:** Choose this when you need structured, iterative deep research using Google's Gemini model and want deterministic outputs with professional report formatting. **When NOT to choose:** Don't choose this if you need multi-model support, as it's locked into Google's Gemini ecosystem and requires an API key. ### [cursor-feedback-extension](https://mcpcatalogs.com/en/server/jianger666-cursor-feedback-extension) by jianger666 · score 43 · ★28 MCP server that enables unlimited AI interactions in Cursor by implementing a feedback loop mechanism within IDE sidebar. **When to choose:** Choose this when working within Cursor IDE and needing to extend AI conversations beyond monthly quota limits, especially for complex tasks requiring iterative feedback. **When NOT to choose:** Don't choose this if you need AI feedback outside of Cursor IDE or prefer a solution with more comprehensive documentation and community support. ### [BrowseAI-Dev](https://mcpcatalogs.com/en/server/browseai-hq-browseai-dev) by BrowseAI-HQ · score 43 · ★17 BrowseAI Dev provides an evidence-backed web search MCP server with citations and confidence scores to reduce AI hallucinations. **When to choose:** Choose BrowseAI-Dev when you need reliable, evidence-backed research with confidence scoring and anti-hallucination verification for AI agents. **When NOT to choose:** Avoid if you need offline access or open-source self-hosting, as BrowseAI-Dev is a commercial service requiring API keys. ### [mcp](https://mcpcatalogs.com/en/server/local-falcon-mcp) by local-falcon · score 43 · ★19 A TypeScript MCP server for Local Falcon's SEO platform with 37 tools for local search tracking and competitor analysis. **When to choose:** Choose this MCP server if you need comprehensive local SEO monitoring and competitor analysis tools integrated into your AI workflow, especially if you're already using Local Falcon's platform. **When NOT to choose:** Don't choose this server if you're not already using or committed to Local Falcon's platform, as it requires an account with the service and is tightly coupled to their API. ### [TranscriptionTools-MCP](https://mcpcatalogs.com/en/server/mushroomfleet-transcriptiontools-mcp) by MushroomFleet · score 43 · ★19 MCP server for transcript processing with repair, formatting, and summarization tools powered by deep-thinking LLMs. **When to choose:** Choose this server for transcript processing tasks where you need to clean up automatic speech recognition output, format meeting transcripts, or create summaries of audio content without needing the original media files. **When NOT to choose:** Don't choose this server if you need to process video files directly or if you require human-level transcription accuracy, as it works on text transcripts rather than audio processing. ### [osmmcp](https://mcpcatalogs.com/en/server/nervsystems-osmmcp) by NERVsystems · score 43 · ★22 OpenStreetMap MCP server providing precision geospatial tools including geocoding, routing, and neighborhood analysis. **When to choose:** Choose osmmcp when you need precise geospatial tools for LLM applications, especially for location-based services, route planning, or neighborhood analysis. **When NOT to choose:** Don't choose osmmcp if you need global coverage outside OpenStreetMap data, or if you require real-time traffic data which isn't provided. ### [codex-agent-mem](https://mcpcatalogs.com/en/server/marcelocaporale-codex-agent-mem) by MarceloCaporale · score 43 · ★21 Local-first MCP memory layer that compresses context packs with SQLite + FTS5 for multiple AI coding workflows. **When to choose:** Choose this when you need persistent local memory for AI agents that must operate offline or with strict privacy requirements, especially for long-running coding workflows. **When NOT to choose:** Not suitable if you need cloud-based memory sharing, real-time multi-user collaboration features, or require write access (this server defaults to read-only). ### [mindmap-mcp-server](https://mcpcatalogs.com/en/server/yuchenssr-mindmap-mcp-server) by YuChenSSR · score 43 · ★231 MCP server that converts Markdown content to interactive mindmaps with HTML or file output options. **When to choose:** Choose this server when you need to convert Markdown content to interactive mindmaps and want options for both HTML preview and file output to save tokens. **When NOT to choose:** Avoid this if you need more advanced mindmap features, customization options beyond basic toolbar visibility, or if you're working on Windows where Docker might be required for proper functionality. ### [agoragentic-integrations](https://mcpcatalogs.com/en/server/rhein1-agoragentic-integrations) by rhein1 · score 43 · ★17 Agoragentic MCP server enables AI agents to buy work from other agents over HTTP with receipts and USDC settlement. **When to choose:** Choose Agoragentic when your agents need to make paid API calls to other services with proper receipt tracking and settlement, especially for cross-agent commerce. **When NOT to choose:** Avoid Agoragentic if you're not ready for USDC settlement, don't need paid service routing, or require a different blockchain for settlement. ### [ytt-mcp](https://mcpcatalogs.com/en/server/cottongeeks-ytt-mcp) by cottongeeks · score 43 · ★71 MCP server for fetching YouTube video transcripts with multiple integrations. **When to choose:** Choose this when you need to analyze YouTube content within your AI workflow, especially if you're already using Claude Desktop or Raycast. **When NOT to choose:** Don't choose this if you need to work with private videos or require processing video content beyond transcript extraction. ### [mcp-prompt-optimizer](https://mcpcatalogs.com/en/server/bubobot-team-mcp-prompt-optimizer) by Bubobot-Team · score 42 · ★23 Advanced MCP server for prompt optimization with research-backed strategies and 11 professional domain templates. **When to choose:** Choose this server when you need research-backed prompt optimization strategies with measurable performance improvements, especially for complex reasoning tasks or enterprise applications requiring domain-specific templates. **When NOT to choose:** Avoid this if you need real-time optimization for extremely high-volume use cases or if you're working with a non-Python ecosystem as this server is Python-based. ### [Devmind-MCP](https://mcpcatalogs.com/en/server/jochenyang-devmind-mcp) by JochenYang · score 42 · ★18 DevMind MCP provides persistent memory capabilities for AI assistants through Model Context Protocol, enabling context retention across conversations. **When to choose:** Choose DevMind MCP when you need a privacy-focused memory system for AI assistants that stores all data locally and provides intelligent context retention across conversations. **When NOT to choose:** Don't choose DevMind MCP if you need cloud-based memory synchronization, multi-user collaboration features, or enterprise-grade security measures. ### [muse-brain](https://mcpcatalogs.com/en/server/falcoschaefer99-eng-muse-brain) by falcoschaefer99-eng · score 42 · ★15 Relational memory substrate for AI agents with 33 MCP tools, identity continuity, and self-hosted persistence. **When to choose:** Choose MUSE Brain when building long-term AI companions or autonomous agents that need persistent identity and memory across sessions, especially if you need sophisticated relational memory with emotional charge and autonomous capabilities. **When NOT to choose:** Don't choose this if you need a simple memory system without the complexity of identity persistence and autonomous execution cycles, or if you're uncomfortable with the CC BY-NC-SA 4.0 license restrictions on commercial use. ### [ggui](https://mcpcatalogs.com/en/server/ggui-ai-ggui) by ggui-ai · score 42 · ★16 ggui is an MCP-UI protocol that generates rich interactive interfaces on demand between AI agents and humans. **When to choose:** Choose ggui when you need AI agents to dynamically generate interactive interfaces without frontend development. **When NOT to choose:** Avoid ggui if you need persistent UIs, complex state management beyond simple updates, or have existing frontend frameworks you want to integrate with. ### [observee](https://mcpcatalogs.com/en/server/observee-ai-observee) by observee-ai · score 42 · ★41 SDK for building AI agents with MCPs, featuring 1000+ integrations, OAuth management and observability tools. **When to choose:** Choose Observee when you need a comprehensive solution for AI agent development with multiple LLM providers, pre-built MCP integrations, and a managed authentication system. **When NOT to choose:** Avoid Observee if you need self-hosted solutions with no external dependencies, or if you prefer a more minimal MCP server without additional observability features. ### [mcp-server-langfuse](https://mcpcatalogs.com/en/server/langfuse-mcp-server-langfuse) by langfuse · score 42 · ★167 MCP server for accessing and managing Langfuse prompts through the Model Context Protocol. **When to choose:** Choose this server if you're already using Langfuse for prompt management and want to integrate those prompts into AI applications that support MCP. **When NOT to choose:** Don't choose this if you're not using Langfuse or need features not supported by this server, such as non-production prompts or detailed argument descriptions. ### [any-chat-completions-mcp](https://mcpcatalogs.com/en/server/pyroprompts-any-chat-completions-mcp) by pyroprompts · score 42 · ★151 MCP server that integrates any OpenAI-compatible API with Claude as a tool for multi-LLM interactions. **When to choose:** Choose this server when you need to integrate Claude with specific OpenAI-compatible LLM APIs not directly supported by Claude, or when you want to use multiple LLM providers within a single Claude session. **When NOT to choose:** Don't choose this if you need specialized features beyond basic chat completion, or if you require non-OpenAI-compatible API endpoints with different request/response formats. ### [codex-claude-bridge](https://mcpcatalogs.com/en/server/abhishekgahlot2-codex-claude-bridge) by abhishekgahlot2 · score 42 · ★38 Bidirectional bridge between Claude Code and OpenAI Codex CLI using MCP and Claude Code Channels. **When to choose:** Choose this when you need to compare or debate between Claude Code and Codex CLI's coding approaches in a unified interface. **When NOT to choose:** Avoid if you need symmetric bidirectional communication as this implementation is asymmetric with better real-time flow from Codex to Claude. ### [buttplug-mcp](https://mcpcatalogs.com/en/server/conacademy-buttplug-mcp) by ConAcademy · score 42 · ★131 MCP server for controlling Buttplug devices through LLM tool calls with vibration and battery status functionality. **When to choose:** Choose this server if you're building intimate applications that need LLM control of sex toys and are comfortable with experimental software. **When NOT to choose:** Avoid this server if you need stable, production-ready device control, as the underlying library has known connection instability issues. ### [uk-case-law-mcp-server](https://mcpcatalogs.com/en/server/georgejeffers-uk-case-law-mcp-server) by georgejeffers · score 42 · ★23 UK case law MCP server using The National Archives API to search, retrieve, and cite UK legal judgments. **When to choose:** Choose this server when you need access to UK legal judgments for legal research, case analysis, or academic purposes with proper citation formatting. **When NOT to choose:** Avoid for commercial use as it's restricted to non-commercial purposes only under PolyForm Noncommercial License. ### [mcp-pinecone](https://mcpcatalogs.com/en/server/sirmews-mcp-pinecone) by sirmews · score 42 · ★148 MCP server enabling Claude Desktop to read/write Pinecone vector databases with semantic search and document processing. **When to choose:** Choose this MCP server when you need to connect Claude Desktop to a Pinecone vector database for RAG applications or semantic search functionality. **When NOT to choose:** Avoid if you need to use other vector databases besides Pinecone, as this server is tightly coupled to Pinecone's API and services. ### [robot_MCP](https://mcpcatalogs.com/en/server/ilialarchenko-robot-mcp) by IliaLarchenko · score 42 · ★78 MCP server for controlling SO-ARM100 robots through LLM agents with manual keyboard control option. **When to choose:** Choose this MCP server if you need to control SO-ARM100 or LeKiwi robots through AI agents with natural language instructions, and especially if you're already using Claude or Gemini models. **When NOT to choose:** Avoid this server if you need support for other robot models, require more recent updates (last commit was 9 months ago), or need a more production-ready solution with long-term maintenance guarantees. ### [fathom-mcp](https://mcpcatalogs.com/en/server/dot-fun-fathom-mcp) by Dot-Fun · score 42 · ★15 MCP server for accessing Fathom AI meeting recordings, transcripts, summaries, teams, and webhooks. **When to choose:** Choose this server if you're already using Fathom AI and want to integrate meeting data into your MCP-enabled workflows or AI assistants. **When NOT to choose:** Don't choose this if you're not using Fathom AI, as it requires their API for functionality and doesn't support other meeting platforms. ### [AgentRunKit](https://mcpcatalogs.com/en/server/tom-ryder-agentrunkit) by Tom-Ryder · score 42 · ★24 Swift 6 agent SDK with type-safe tools, streaming support, and MCP integration. **When to choose:** Developers working on Swift applications who need type-safe LLM integration with MCP support, especially for on-device inference on Apple platforms. **When NOT to choose:** Teams not using Swift or Apple platforms, or those needing cross-platform support beyond iOS/macOS. ### [chatgpt2md](https://mcpcatalogs.com/en/server/nextstat-chatgpt2md) by NextStat · score 42 · ★16 ChatGPT to Markdown converter with full-text search and MCP server for accessing conversations in Claude. **When to choose:** Choose this tool if you're migrating from ChatGPT to Claude and need to maintain access to your conversation history with full search capabilities. **When NOT to choose:** Don't choose this if you need to process exports from platforms other than ChatGPT or if you require bidirectional sync between ChatGPT and Claude. ### [ultrathink](https://mcpcatalogs.com/en/server/husniadil-ultrathink) by husniadil · score 42 · ★16 UltraThink MCP server enables structured sequential thinking with confidence scoring, assumption tracking, and multi-session support. **When to choose:** Choose UltraThink when you need structured sequential thinking with explicit confidence scoring and assumption tracking for complex problem-solving. **When NOT to choose:** Avoid if you need persistent sessions across server restarts, as UltraThink only stores sessions in memory and they're lost when the server terminates. ### [knowledgebase-mcp](https://mcpcatalogs.com/en/server/biocontext-ai-knowledgebase-mcp) by biocontext-ai · score 42 · ★23 BioContextAI Knowledgebase MCP server provides unified access to biomedical APIs for AI agents. **When to choose:** Choose this MCP server when building biomedical research AI systems that need access to multiple specialized databases and literature sources through a unified interface. **When NOT to choose:** Don't choose this if you need offline-only access, as many tools require internet connections to biomedical APIs, and avoid for commercial applications without checking individual data source licenses. ### [FastAPI-BitNet](https://mcpcatalogs.com/en/server/grctest-fastapi-bitnet) by grctest · score 42 · ★38 FastAPI-based MCP server for Microsoft's BitNet model with session management, chat, and benchmarking capabilities. **When to choose:** Choose FastAPI-BitNet if you need an MCP interface for Microsoft's BitNet model with comprehensive session management and benchmarking capabilities, particularly when integrating with VS Code. **When NOT to choose:** Not suitable for production environments requiring high availability or load balancing, as it appears to be a single-instance implementation without built-in redundancy. ### [programmatic-tool-calling-ai-sdk](https://mcpcatalogs.com/en/server/cameronking4-programmatic-tool-calling-ai-sdk) by cameronking4 · score 42 · ★17 An MCP-enabled SDK that reduces LLM costs 80% by replacing tool calls with generated JavaScript executed in Vercel Sandbox. **When to choose:** Choose this when you need to optimize costs and performance for applications requiring multiple tool calls, especially when working with both local tools and MCP servers. **When NOT to choose:** Don't choose this if you can't use Vercel services (Sandbox requires Vercel account) or if you need strict tool execution guarantees rather than code generation. ### [mcp-memory](https://mcpcatalogs.com/en/server/puliczek-mcp-memory) by Puliczek · score 42 · ★143 MCP Memory server that enables cross-conversation user memory using vector search technology on Cloudflare infrastructure. **When to choose:** Choose this MCP Memory server when you need persistent user memory across conversations for personalized AI interactions, especially if already using Cloudflare services. **When NOT to choose:** Avoid this if you need on-premises hosting, require different vector database providers, or want to avoid Cloudflare vendor lock-in. ### [mcp-summarization-functions](https://mcpcatalogs.com/en/server/braffolk-mcp-summarization-functions) by Braffolk · score 41 · ★37 MCP server providing intelligent text summarization for AI agents to optimize context usage. **When to choose:** Choose this server when working with AI agents that process large outputs, file contents, or directory structures to prevent context window overflow and improve reliability. **When NOT to choose:** Avoid if you need access to raw file content without summarization, or if you're concerned about API costs from additional model calls for summarization. ### [Remote-MCP](https://mcpcatalogs.com/en/server/ssut-remote-mcp) by ssut · score 41 · ★206 Type-safe remote MCP communication solution enabling centralized management of model contexts. **When to choose:** Choose this when you need immediate remote access to MCP servers before official implementations are available, and when you want centralized management of distributed MCP resources. **When NOT to choose:** Avoid this if you need production-ready stability, as it's marked as experimental with potential breaking changes, or if you prefer to wait for official MCP remote support. ### [protocols-io-mcp-server](https://mcpcatalogs.com/en/server/hqn21-protocols-io-mcp-server) by hqn21 · score 41 · ★16 A functional MCP server enabling Claude Desktop to interact with protocols.io scientific protocols. **When to choose:** Choose this MCP server if you work with scientific protocols and want to integrate protocol management into your AI workflow. **When NOT to choose:** Don't choose this if you need protocol management outside of the protocols.io ecosystem or require OAuth with a custom identity provider. ### [mentor-mcp-server](https://mcpcatalogs.com/en/server/cyanheads-mentor-mcp-server) by cyanheads · score 41 · ★31 MCP server providing LLM agents with expert mentorship for code review, design critique, and writing feedback. **When to choose:** Choose this server when your LLM agents need expert second opinions for code, design, writing, or feature planning, especially if you already have access to Deepseek API. **When NOT to choose:** Don't choose this if you don't have access to Deepseek API or need a completely offline solution, as this server requires external API calls. ### [mcp-gemini-server](https://mcpcatalogs.com/en/server/bsmi021-mcp-gemini-server) by bsmi021 · score 41 · ★35 A TypeScript MCP server wrapping Google Gemini API, offering tools for text generation, multimedia analysis, and function calling. **When to choose:** Choose this server when you need to integrate Google's Gemini models into an MCP-compatible system, especially for multimedia analysis from URLs and image generation capabilities. **When NOT to choose:** Don't choose this if you need direct file upload support or want to use Vertex AI credentials instead of Google AI Studio API keys. ### [nakkas](https://mcpcatalogs.com/en/server/arikusi-nakkas) by arikusi · score 41 · ★8 MCP server that turns AI into an SVG artist, generating animated SVG graphics from JSON configuration. **When to choose:** Choose Nakkas when you need to generate complex animated SVG graphics through AI without learning SVG syntax, especially for web visuals, data visualization, or design prototyping. **When NOT to choose:** Don't choose Nakkas if you need interactive SVG elements with event handlers (onclick) or require advanced text features like custom font loading in static environments. ### [qurio](https://mcpcatalogs.com/en/server/irahardianto-qurio) by irahardianto · score 41 · ★16 A self-hosted RAG engine for AI coding assistants that ingests technical docs and code repositories locally, serving grounded context via MCP to prevent hallucinations. **When to choose:** Choose Qurio when you need a self-hosted RAG solution for AI coding assistants that respects your privacy and processes technical documentation with structure-aware chunking. **When NOT to choose:** Don't choose Qurio if you need cloud-based deployment or require authentication/authorization for multi-user access as it's designed for localhost-only usage. ### [MCP-wolfram-alpha](https://mcpcatalogs.com/en/server/secretiveshell-mcp-wolfram-alpha) by SecretiveShell · score 40 · ★75 MCP server connecting to Wolfram Alpha API for computational intelligence and queries. **When to choose:** Choose this when you need to integrate computational intelligence and mathematical calculations into chat-based AI systems. **When NOT to choose:** Avoid if you need free API access or if Wolfram Alpha doesn't cover your specific domain knowledge requirements. ### [mcp_espn_ff](https://mcpcatalogs.com/en/server/kbthree13-mcp-espn-ff) by KBThree13 · score 40 · ★30 ESPN Fantasy Football MCP server providing league data, team rosters, and player stats through tools. **When to choose:** Choose this MCP server if you use ESPN Fantasy Football and want to access league data through natural language interactions with LLMs. **When NOT to choose:** Avoid if you use other fantasy football platforms (like Yahoo or NFL) that don't have ESPN API integration. ### [ComIDP-MCP-Server](https://mcpcatalogs.com/en/server/compdfkit-comidp-mcp-server) by ComPDFKit · score 40 · ★83 ComIDP MCP Server extracts data from PDFs and integrates with Claude Desktop via Model Context Protocol. **When to choose:** Teams already using ComPDFKit's services who need to extract PDF data through Claude Desktop without additional infrastructure. **When NOT to choose:** You need document processing for formats other than PDF, or you don't want vendor lock-in to ComPDFKit's services. ### [pophive-mcp-server](https://mcpcatalogs.com/en/server/cicatriiz-pophive-mcp-server) by Cicatriiz · score 40 · ★20 MCP server providing access to Yale's PopHIVE public health data with comprehensive tools for health surveillance analysis. **When to choose:** Choose this MCP server when you need access to near real-time public health data from authoritative sources including CDC surveillance systems, Epic Cosmos EHR networks, and Google Health Trends for epidemiological research or public health analysis. **When NOT to choose:** Avoid this server if you need international health data, as it currently only covers US public health data, or if you require custom data sources not already available through PopHIVE. ### [cursor10x-mcp](https://mcpcatalogs.com/en/server/aiurda-cursor10x-mcp) by aiurda · score 40 · ★79 DevContext (formerly Cursor10x) is an MCP server providing persistent multi-dimensional memory for AI assistants. **When to choose:** Choose this if you're using Cursor/Claude and need persistent context across development sessions, especially for complex projects with many code relationships. **When NOT to choose:** Don't choose this if you're not using Cursor/Claude, if you prefer minimal resource usage, or if you're concerned about vendor lock-in with Turso. ### [perplexity-mcp-server](https://mcpcatalogs.com/en/server/cyanheads-perplexity-mcp-server) by cyanheads · score 40 · ★22 A TypeScript MCP server that provides tools for Perplexity API search and deep research capabilities. **When to choose:** Choose this when you need your AI agents to access real-time web search and comprehensive research capabilities through a standardized MCP interface, especially for tasks requiring recent information or multi-source validation. **When NOT to choose:** Avoid if you require offline capabilities, need support for search providers other than Perplexity, or prefer self-hosted solutions without third-party API dependencies. ### [aminer-mcp](https://mcpcatalogs.com/en/server/huanghuoguoguo-aminer-mcp) by huanghuoguoguo · score 40 · ★6 AMiner MCP server provides free access to academic databases (scholars, papers, patents) via the Model Context Protocol. **When to choose:** Choose this MCP server when you need cost-free academic data retrieval for AI applications, especially for research assistants, literature reviews, or patent analysis. **When NOT to choose:** Avoid if you need real-time academic data updates, as free access may be limited or change without notice, or if you require features beyond the current five search tools. ### [archiledger](https://mcpcatalogs.com/en/server/thecookiezen-archiledger) by thecookiezen · score 40 · ★5 Archiledger is a graph-based MCP memory server with vector search capabilities for persistent AI knowledge storage. **When to choose:** Choose Archiledger when you need persistent, graph-structured memory for AI assistants and want both manual control and AI-powered automation capabilities. **When NOT to choose:** Don't choose this for production use as it lacks authentication, or if you prefer cloud-based solutions over local embedded databases. ### [arxiv-mcp-server](https://mcpcatalogs.com/en/server/anuj0456-arxiv-mcp-server) by anuj0456 · score 40 · ★18 MCP server for arXiv enabling AI assistants to search, retrieve, analyze, and export academic papers. **When to choose:** Choose this server when you need AI assistants to interact with academic papers from arXiv, especially for literature reviews, trend analysis, or research paper summarization. **When NOT to choose:** Don't choose this if you need access to other academic databases beyond arXiv or require more advanced citation tracking beyond estimated metrics. ### [seedream-image-mcp](https://mcpcatalogs.com/en/server/wearzdk-seedream-image-mcp) by wearzdk · score 40 · ★18 An MCP server integrating SeeDream AI image generation into various AI clients like Cursor and Claude. **When to choose:** Choose this MCP server when you need integrated AI image generation in AI coding assistants like Cursor or Claude Code and are already using Volcengine's SeeDream service. **When NOT to choose:** Avoid this if you need permanent image storage without expiration, or if you're not comfortable with Volcengine's vendor lock-in for image generation. ### [paperclip](https://mcpcatalogs.com/en/server/matsjfunke-paperclip) by matsjfunke · score 40 · ★27 MCP server enabling search and retrieval of research papers from Arxiv, OSF API & OpenAlex. **When to choose:** Choose Paperclip when you need to search academic papers across multiple repositories through an AI assistant interface. **When NOT to choose:** Don't choose Paperclip if you need full-text paper access (only metadata retrieval) or prefer a hosted service without self-hosting. ### [mcp-chain-of-draft-prompt-tool](https://mcpcatalogs.com/en/server/brendancopley-mcp-chain-of-draft-prompt-tool) by brendancopley · score 40 · ★19 An MCP server implementing Chain-of-Draft (CoD) reasoning to enhance LLM responses with reduced token usage. **When to choose:** Choose this tool when you need enhanced reasoning quality with significantly reduced token costs, especially for tasks where CoT approaches are too expensive or slow. **When NOT to choose:** Don't choose this if you need extremely high accuracy for complex tasks where CoD's concise approach might lack sufficient detail, or if you prefer a complete out-of-box solution without configuration steps. ### [PIXRA](https://mcpcatalogs.com/en/server/dodufish-pixra) by dodufish · score 40 · ★239 PIXRA is a reliability-focused framework with MCP server integration for agent workflows and automation. **When to choose:** Choose PIXRA when you need high-reliability agent workflows with verification mechanisms and want to leverage existing MCP tools without custom development. **When NOT to choose:** Avoid PIXRA if you need a framework purely in Java (examples are Python), if you require heavy customization of reliability mechanisms, or if you need minimal agent infrastructure without complex verification layers. ### [claude-mcp-scheduler](https://mcpcatalogs.com/en/server/tonybentley-claude-mcp-scheduler) by tonybentley · score 39 · ★16 A cron-based scheduler that uses Claude API with MCP servers for tool calls in scheduled AI tasks. **When to choose:** Choose this if you need to schedule regular AI tasks on servers or in CI/CD pipelines and want to leverage MCP servers for context-aware tool execution. **When NOT to choose:** Avoid this if you need interactive AI assistance or real-time conversation, as it's designed for scheduled tasks rather than interactive sessions. ### [mcp-gemini-search](https://mcpcatalogs.com/en/server/arjunprabhulal-mcp-gemini-search) by arjunprabhulal · score 39 · ★56 MCP server integrating Gemini 2.5 Pro with flight search tools via function calling. **When to choose:** Choose this MCP server if you need natural language flight search capabilities using Gemini and want to integrate it with other MCP-compatible tools. **When NOT to choose:** Avoid this if you need alternative flight providers beyond SerpAPI or prefer solutions with more extensive tool coverage beyond just flight searches. ### [python-mcp-server-client](https://mcpcatalogs.com/en/server/gobinfan-python-mcp-server-client) by GobinFan · score 39 · ★155 A MCP server for querying technical documentation of major AI agent frameworks like LangChain and LlamaIndex. **When to choose:** Choose this MCP server when you need quick access to documentation for popular AI frameworks without leaving your development environment. **When NOT to choose:** Avoid this if you need offline access to documentation or require more reliable parsing methods than web scraping. ### [paper-search-mcp-openai](https://mcpcatalogs.com/en/server/adamamer20-paper-search-mcp-openai) by adamamer20 · score 39 · ★13 Python-based MCP server for searching and downloading academic papers from multiple sources. **When to choose:** Choose this server if you need academic paper search functionality integrated into your LLM workflow, especially if you're already using Claude Desktop. **When NOT to choose:** Don't choose this if you need access to subscription-only research databases or require more extensive paper management features beyond search and download. ### [strale](https://mcpcatalogs.com/en/server/strale-io-strale) by strale-io · score 39 · ★3 Strale provides MCP server with 290+ quality-scored AI agent capabilities including compliance checks, financial validation and Web3 security. **When to choose:** Choose Strale when building AI agents that need reliable, validated access to financial and compliance data with measurable quality scoring. **When NOT to choose:** Avoid Strale if you need self-hosted capabilities, have strict data residency requirements, or prefer not to use a proprietary service with vendor lock-in. ### [arxiv-mcp-server](https://mcpcatalogs.com/en/server/1dark134-arxiv-mcp-server) by 1Dark134 · score 39 · ★16 arXiv MCP server enabling AI assistants to search, retrieve, analyze, and summarize academic papers with advanced features. **When to choose:** Choose this MCP server for AI assistants that need comprehensive arXiv paper discovery, citation analysis, and academic research workflows with multiple export formats. **When NOT to choose:** Don't choose this if you need real-time access to the latest papers (it may have ingestion delays) or if you require write capabilities to the arXiv database (this server is read-only). ### [rag-anythink-mcp](https://mcpcatalogs.com/en/server/serkanyasr-rag-anythink-mcp) by serkanyasr · score 39 · ★6 MCP server for advanced RAG with knowledge graphs, document processing, and multimodal AI support. **When to choose:** Choose this server when you need advanced document analysis with knowledge graph relationships and multimodal content extraction, especially if you're already using Neo4j and PostgreSQL. **When NOT to choose:** Avoid this if you need a lightweight solution or don't want to manage Neo4j and PostgreSQL infrastructure, as it requires substantial setup resources. ### [agent-board](https://mcpcatalogs.com/en/server/quentintou-agent-board) by quentintou · score 39 · ★21 Open-source multi-agent task board with Kanban, DAG dependencies, and MCP server integration. **When to choose:** Choose Agent Board when you need to coordinate multiple AI agents working on interdependent tasks, especially if you're already using OpenClaw or building complex multi-step workflows. **When NOT to choose:** Not ideal for simple single-agent workflows or if you need database persistence (Agent Board uses file-based storage). ### [gemsuite-mcp](https://mcpcatalogs.com/en/server/pv-bhat-gemsuite-mcp) by PV-Bhat · score 38 · ★29 Professional Gemini API integration for MCP hosts with intelligent model selection and advanced file handling. **When to choose:** Choose GemSuite MCP when you need advanced Gemini API integration with intelligent model selection and efficient file processing for MCP-compatible hosts like Claude. **When NOT to choose:** Avoid this if you need model selection outside Gemini's family, require fine-grained control over model parameters, or need to work offline without API access. ### [adk-python-mcp-client](https://mcpcatalogs.com/en/server/arjunprabhulal-adk-python-mcp-client) by arjunprabhulal · score 38 · ★34 Python implementation of ADK as an MCP client for flight search with Gemini 2 LLM. **When to choose:** Choose this implementation for learning how to build MCP clients with Google's ADK, particularly for those already using Google Cloud services and Gemini models. **When NOT to choose:** Not ideal for production systems requiring vendor diversity or those not using Google's ecosystem, as it ties the implementation to Google-specific technologies. ### [cfbd-mcp-server](https://mcpcatalogs.com/en/server/lenwood-cfbd-mcp-server) by lenwood · score 38 · ★26 CFBD MCP server provides college football statistics API access within Claude Desktop. **When to choose:** Choose this MCP server if you need to query college football data within Claude Desktop and want natural language access to game statistics, team records, and analytics. **When NOT to choose:** Avoid this server if you need data for sports other than college football, as it only specializes in NCAA football statistics. ### [converse](https://mcpcatalogs.com/en/server/falldownthesystem-converse) by FallDownTheSystem · score 38 · ★2 MCP server enabling Claude to converse with multiple AI models through a unified interface. **When to choose:** Choose this when you need to compare responses across multiple AI providers or get consensus from several models on complex questions. **When NOT to choose:** Avoid this if you're concerned about API costs, as running multiple models simultaneously can be expensive, or if you need specialized features not offered by the supported providers. ### [mingli-mcp](https://mcpcatalogs.com/en/server/spyfree-mingli-mcp) by spyfree · score 38 · ★2 A feature-rich fortune telling MCP server supporting Zi Wei Doushu, Bazi, and planned astrology systems with multi-language output. **When to choose:** Choose this server if you need Chinese fortune-telling integration in AI applications, especially for personal insights based on traditional systems like Ziwei or Bazi. **When NOT to choose:** Not suitable if you need Western astrology, comprehensive historical data, or analysis outside traditional Chinese metaphysical systems. ### [mcp-meme-sticky](https://mcpcatalogs.com/en/server/nkapila6-mcp-meme-sticky) by nkapila6 · score 38 · ★13 Python MCP server for creating AI-generated memes and converting them to Telegram stickers. **When to choose:** Choose this server when you want to create custom memes and convert them to Telegram stickers without API keys and with seamless integration through MCP tool calling. **When NOT to choose:** Don't choose this server if you need WhatsApp sticker conversion (still pending) or require more meme generation features beyond what's currently available. ### [ReasoningBank-MCP](https://mcpcatalogs.com/en/server/hanw39-reasoningbank-mcp) by hanw39 · score 38 · ★9 A production-ready MCP server implementing ReasoningBank: Memory as Test-Time Compute Scaling for AI agents to evolve reasoning capabilities through experience. **When to choose:** Choose this server for AI agents that need to learn from experience and improve their reasoning over time, especially in multi-agent systems where specialized roles need isolated memory domains. **When NOT to choose:** Don't choose this if you need immediate results from every interaction without memory retention, or if you're working with extremely simple tasks that don't benefit from accumulated experience. ### [MCPNotes](https://mcpcatalogs.com/en/server/9ninety-mcpnotes) by 9Ninety · score 38 · ★36 A dual-server MCP system for AI-powered note-taking with web interface and DynamoDB storage. **When to choose:** Choose this MCP server if you need AI-powered note management and already use AWS services, as it provides both MCP integration and a web interface. **When NOT to choose:** Avoid this server if you don't want to set up AWS infrastructure or need a self-hosted solution without cloud dependencies. ### [unitree-go2-mcp-server](https://mcpcatalogs.com/en/server/lpigeon-unitree-go2-mcp-server) by lpigeon · score 37 · ★78 MCP server that translates natural language commands to ROS2 instructions for controlling Unitree Go2 robot. **When to choose:** Choose this MCP server if you have a Unitree Go2 robot and want to control it using natural language commands through an LLM interface. **When NOT to choose:** Don't choose this if you don't have a Unitree Go2 robot, as it's specifically designed for this hardware platform. ### [mcp-copilot](https://mcpcatalogs.com/en/server/tshu-w-mcp-copilot) by tshu-w · score 37 · ★18 A meta MCP server that automatically routes queries to 1000+ MCP servers without overwhelming the LLM. **When to choose:** When you need to manage access to hundreds of MCP servers without overwhelming the LLM context window, or when building scalable AI systems that need to orchestrate many specialized tools. **When NOT to choose:** If you only need to work with a small number of MCP servers (<10), or if you require tools that aren't available in the routed pool of servers. ### [ollama-mcp-example](https://mcpcatalogs.com/en/server/kirillsaidov-ollama-mcp-example) by kirillsaidov · score 37 · ★18 A beginner-friendly example demonstrating MCP server and client implementation with Ollama. **When to choose:** Choose this when you're learning MCP basics or need a simple, minimal example to understand how MCP servers and clients work together. **When NOT to choose:** Don't choose this for production use cases as it's an example with minimal error handling and no authentication mechanisms. ### [quick-mcp-example](https://mcpcatalogs.com/en/server/alucek-quick-mcp-example) by ALucek · score 37 · ★60 Minimal example MCP server implementation demonstrating Tools, Resources and Prompts capabilities in Python. **When to choose:** Choose this example when you're learning MCP implementation patterns or need a starting template for building a knowledgebase chatbot with RAG capabilities. **When NOT to choose:** Don't choose this for production use as it's an educational example rather than a fully-featured server. For production systems, consider more mature MCP servers with authentication, error handling, and support. ### [mcpet](https://mcpcatalogs.com/en/server/shreyaskarnik-mcpet) by shreyaskarnik · score 37 · ★10 MCPet is a TypeScript-based virtual pet simulation MCP server with interactive pet care tools. **When to choose:** Choose MCPet for entertainment and educational purposes when implementing or learning MCP server functionality, particularly for simple interactive applications with visual feedback. **When NOT to choose:** Don't choose MCPet for serious applications or environments requiring persistent data, as it's designed primarily as a demonstration/entertainment tool with limited data persistence. ### [mcp-llm](https://mcpcatalogs.com/en/server/sammcj-mcp-llm) by sammcj · score 37 · ★78 An MCP server providing LLMs access to other LLMs with code generation, documentation creation, and question answering tools. **When to choose:** Choose this server when you need programmatic access to LLM capabilities for code generation and documentation within an MCP workflow. **When NOT to choose:** Avoid this if you need enterprise-grade support or more advanced features like model customization, as this appears to be a smaller project with limited documentation. ### [mcp-server-amazon-bedrock](https://mcpcatalogs.com/en/server/zxkane-mcp-server-amazon-bedrock) by zxkane · score 37 · ★24 MCP server for Amazon Bedrock Nova Canvas AI image generation with configurable parameters. **When to choose:** Choose this server if you already use AWS services and need high-quality image generation through Amazon Bedrock's Nova Canvas model with MCP integration. **When NOT to choose:** Avoid if you don't want AWS vendor lock-in or if you need image generation services outside of AWS ecosystem. ### [mcp-chatbot](https://mcpcatalogs.com/en/server/mctrinh-mcp-chatbot) by mctrinh · score 37 · ★17 MCP-powered research assistant using Claude 3 for literature search and summarization. **When to choose:** Choose this if you're an academic or engineer who needs to quickly search and summarize research literature using Claude 3. **When NOT to choose:** Avoid if you need access to other LLM providers besides Anthropic, or if you require a web UI interface (currently planned but not implemented). ### [jiki](https://mcpcatalogs.com/en/server/teilomillet-jiki) by teilomillet · score 37 · ★17 Jiki is a Python framework that connects LLMs to external tools via MCP, offering both orchestration and client capabilities. **When to choose:** Choose Jiki if you're building Python applications that need to connect LLMs with external tools via MCP and want a flexible orchestration layer. **When NOT to choose:** Avoid Jiki if you need non-Python tool servers, as it currently only integrates with MCP servers in Python or requires custom implementation. ### [vibe-marketing](https://mcpcatalogs.com/en/server/synthetic-ci-vibe-marketing) by synthetic-ci · score 37 · ★6 Marketing-focused MCP server providing hooks, copywriting frameworks, and content optimization tools for social media. **When to choose:** Choose this server if you need specialized marketing content tools and want to stay updated with the latest social media trends. **When NOT to choose:** Don't choose this if you need deep analytics or full-fledged social media management, as it focuses primarily on content creation assistance. ### [remember-mcp-vscode](https://mcpcatalogs.com/en/server/niclasolofsson-remember-mcp-vscode) by NiclasOlofsson · score 37 · ★5 VS Code extension that brings persistent memory to your AI assistant through MCP integration. **When to choose:** Choose this extension if you use VS Code with GitHub Copilot and want persistent memory for personal preferences and team knowledge without additional setup. **When NOT to choose:** Avoid this if you're not using VS Code or GitHub Copilot, or if you need more advanced memory management features beyond what the Mode Manager MCP server provides. ### [go-mcp-server](https://mcpcatalogs.com/en/server/cbrgm-go-mcp-server) by cbrgm · score 37 · ★6 A learning MCP server in Go that implements tea-related tools, resources, and prompts with stdio and HTTP transport options. **When to choose:** Go developers looking for a complete example of MCP implementation for learning purposes or as a template for domain-specific servers. **When NOT to choose:** For production use cases, as the author explicitly states this is for learning only and recommends using mcp-go or the official MCP Go SDK instead. ### [ga4-mcp-server](https://mcpcatalogs.com/en/server/spindle79-ga4-mcp-server) by spindle79 · score 36 · ★0 Google Analytics 4 MCP server exposing metrics to AI agents with daily/monthly breakdowns and dual MCP/REST endpoints. **When to choose:** Choose this when you need to integrate Google Analytics 4 data into AI systems and want both MCP and REST access options with proper time breakdowns. **When NOT to choose:** Avoid this if you need access to Universal Analytics data, require real-time analytics with sub-minute granularity, or need more than 5 concurrent API calls to Google. ### [mcp-flight-search](https://mcpcatalogs.com/en/server/arjunprabhulal-mcp-flight-search) by arjunprabhulal · score 36 · ★42 Python MCP server for flight search using SerpAPI, providing tools to search for flights and check server status. **When to choose:** Choose this MCP server if you need to integrate real-time flight search capabilities into an AI application that uses the Model Context Protocol. **When NOT to choose:** Don't choose this if you need flight booking capabilities (search-only), have API quota constraints with SerpAPI, or require offline functionality. ### [sarvam-mcp-server](https://mcpcatalogs.com/en/server/shobhit-nagpal-sarvam-mcp-server) by Shobhit-Nagpal · score 36 · ★12 MCP server for Sarvam API integration, providing language processing tools for Indian languages. **When to choose:** Choose this server when working with Indian languages and need text processing capabilities without writing code. **When NOT to choose:** Avoid if you need language support outside of Indian languages or require features not provided by the Sarvam API. ### [jgrants-mcp](https://mcpcatalogs.com/en/server/tachibanayu24-jgrants-mcp) by tachibanayu24 · score 36 · ★3 An MCP server that provides access to Japanese subsidy information through the jGrants API. **When to choose:** Choose this server if you need Japanese-language access to government subsidy information and want to integrate it into AI workflows through MCP. **When NOT to choose:** Don't choose this server if you need subsidy information outside Japan, as it's specifically designed for Japanese grants only. ### [gen-ai-starter](https://mcpcatalogs.com/en/server/jbsoftware-io-gen-ai-starter) by jbsoftware-io · score 36 · ★3 Generative AI starter with MCP integration for converting REST APIs to standardized tools. **When to choose:** Choose this as a learning resource for understanding how to convert REST APIs to MCP tools or when you need a working example of MCP server implementation. **When NOT to choose:** Not suitable for production use as this is primarily a starter project with a single API example (PokéAPI) rather than a comprehensive MCP server framework. ### [seoul-essentials](https://mcpcatalogs.com/en/server/do-droid-seoul-essentials) by do-droid · score 36 · ★0 An MCP server providing Seoul's public facility data for tourist assistance AI agents. **When to choose:** Choose Seoul Essentials when building AI assistants for foreign tourists in Seoul who need real-time information about public facilities, healthcare resources, and transportation options. **When NOT to choose:** Don't choose this if you need extensive geographic coverage beyond Seoul, require real-time data updates, or need integration with non-Korean public services. ### [ai-compliance-monitor](https://mcpcatalogs.com/en/server/vdineshk-ai-compliance-monitor) by vdineshk · score 36 · ★0 MCP server providing structured regulatory intelligence for AI agents to ensure compliance across jurisdictions. **When to choose:** Teams building AI agents that need to comply with multiple international regulations, especially those already using or comfortable with Cloudflare infrastructure. **When NOT to choose:** If you need compliance coverage for jurisdictions beyond EU, Singapore, and Colorado, or if you require self-hosting options outside of Cloudflare. ### [LLaMa-MCP-Streamlit](https://mcpcatalogs.com/en/server/nikunj2003-llama-mcp-streamlit) by Nikunj2003 · score 36 · ★43 A Streamlit AI assistant using MCP to enable tool interactions with LLaMa 3.3 or Ollama models. **When to choose:** Choose this when you need a quick prototype for an MCP-integrated AI assistant with a UI, especially if you're already using Streamlit or Docker. **When NOT to choose:** Don't choose this for production deployment due to limited documentation and the project appears to be a prototype rather than a fully-fledged application. ### [aiskillstore](https://mcpcatalogs.com/en/server/garasegae-aiskillstore) by garasegae · score 36 · ★0 MCP server for AI agents to discover, install, and review skills in the agent-first skill marketplace. **When to choose:** Choose this MCP server when you need AI agents to discover, evaluate, and install skills across multiple platforms with trust validation and schema inspection. **When NOT to choose:** Avoid this if you're building a closed ecosystem with proprietary skills, or if you need offline access to skills without dependency on external services. ### [websimm](https://mcpcatalogs.com/en/server/gigachadtrey-websimm) by gigachadtrey · score 36 · ★0 Comprehensive MCP server for WebSim API, providing access to projects, users, feeds, assets, and comments with robust pagination and error handling. **When to choose:** Choose this MCP server if you're working with WebSim's platform and need programmatic access to projects, users, and content through standardized MCP tools. **When NOT to choose:** Not suitable for applications requiring write access to WebSim data as this server only provides read-only access to the public API. ### [eve-online-traffic-mcp](https://mcpcatalogs.com/en/server/kongyo2-eve-online-traffic-mcp) by kongyo2 · score 36 · ★4 An MCP server for EVE Online traffic, navigation, and system information using both the official ESI API and SDE data. **When to choose:** Choose this MCP server if you're working with EVE Online data, need route planning between systems, or want to analyze traffic patterns in the EVE universe. **When NOT to choose:** Don't choose this if you need access to character-specific private data, as this server only provides public universe information and doesn't support authentication. ### [loopin-mcp](https://mcpcatalogs.com/en/server/nicholasemccormick-loopin-mcp) by nicholasemccormick · score 36 · ★0 LoopIn is an MCP server providing human-in-the-loop interrupt API for AI agents to request human review at critical decision points. **When to choose:** Choose LoopIn when building AI agents that require human oversight at critical decision points, especially for financial transactions or system operations. **When NOT to choose:** Avoid LoopIn if you need real-time, collaborative decision-making with multiple humans or require workflow integration beyond simple approval/rejection. ### [mcp-server-ga4](https://mcpcatalogs.com/en/server/harshfolio-mcp-server-ga4) by harshfolio · score 35 · ★19 An MCP server enabling LLMs to access and analyze Google Analytics 4 data through GA4 Data API. **When to choose:** Choose this if you need to integrate Google Analytics 4 data into AI applications through MCP protocol and already have GA4 set up with proper authentication. **When NOT to choose:** Don't choose this if you need GA4 data from multiple accounts or require write access to GA4 (this server is read-only only). ### [mcp-server-bwt](https://mcpcatalogs.com/en/server/zizzfizzix-mcp-server-bwt) by zizzfizzix · score 34 · ★6 Python MCP server exposing Bing Webmaster Tools API functionality to AI assistants like Claude. **When to choose:** Choose this MCP server if you work with Bing Webmaster Tools regularly and want to streamline SEO tasks through conversational AI interactions. **When NOT to choose:** Don't choose this if you primarily work with other search engines' webmaster tools or need functionality not exposed by the Bing Webmaster API. ### [kaggle-MCP](https://mcpcatalogs.com/en/server/dishant27-kaggle-mcp) by Dishant27 · score 34 · ★9 MCP server enabling AI assistants to interact with Kaggle competitions through commands and API access. **When to choose:** Choose this server if you're a data scientist or ML practitioner who wants to integrate Kaggle competitions directly into your AI workflow without switching contexts between the Kaggle website and your AI assistant. **When NOT to choose:** Don't choose this if you need access to private competitions (requires manual acceptance of terms on Kaggle) or if you're already committed to another data science platform with built-in AI integration. ### [luma-api-mcp](https://mcpcatalogs.com/en/server/lumalabs-luma-api-mcp) by lumalabs · score 34 · ★24 An MCP server that integrates Luma AI's image and video generation capabilities through Ray and Photon models. **When to choose:** Choose this MCP server when you need direct access to Luma AI's image and video generation capabilities through Claude Desktop or other MCP clients. **When NOT to choose:** Avoid if you need generation capabilities outside of Luma AI's models or require higher performance than the API provides. ### [CereBro](https://mcpcatalogs.com/en/server/rob1997-cerebro) by rob1997 · score 34 · ★17 A .NET MCP client-server implementation for AI models, currently supporting OpenAI with plans for other models. **When to choose:** Choose CereBro if you're building .NET applications or Unity games that need MCP integration and want a model-agnostic approach to AI tool calling. **When NOT to choose:** Avoid CereBro if you need simultaneous multi-model support or are working outside the .NET/Unity ecosystem where it would require significant adaptation. ### [wwo-mcp](https://mcpcatalogs.com/en/server/worldweatheronline-wwo-mcp) by WorldWeatherOnline · score 34 · ★1 World Weather Online MCP server provides comprehensive weather data tools for AI assistants. **When to choose:** Choose this MCP server when you need comprehensive weather data for AI applications and prefer a ready-to-deploy solution with detailed configuration instructions. **When NOT to choose:** Don't choose this if you need weather data from a different provider or if you want to avoid third-party API dependencies in your AI application. ### [ai-agent-mcp](https://mcpcatalogs.com/en/server/alexey-tyurin-ai-agent-mcp) by alexey-tyurin · score 33 · ★4 AI content moderation system using OpenAI's API through Google's ADK and MCP protocol. **When to choose:** Choose this MCP server when you need to integrate content moderation into AI applications using Google ADK with OpenAI's API. **When NOT to choose:** Don't choose this if you need an alternative moderation API provider or require authentication beyond API keys. ### [mcp-carbon-calculator](https://mcpcatalogs.com/en/server/prdai-mcp-carbon-calculator) by prdai · score 33 · ★0 Australian carbon emissions calculator MCP server using official NGA 2024 data with TypeScript implementation. **When to choose:** Choose this MCP server when you need accurate carbon emission calculations specifically for Australian electricity and gas consumption, particularly for environmental reporting, sustainability assessments, or carbon footprint tracking in the Australian context. **When NOT to choose:** Don't choose this if you need carbon calculations for regions outside of Australia, as the emission factors are specific to the Australian National Greenhouse Accounts. ### [MCPGateway](https://mcpcatalogs.com/en/server/pokshiu-mcpgateway) by pokshiu · score 33 · ★0 MCPGateway unifies local AITools and remote MCP servers into one searchable execution surface using .NET 10. **When to choose:** Windows users who want a simplified GUI interface for discovering and running various AI tools without technical knowledge. **When NOT to choose:** Users on non-Windows platforms, developers needing programmatic access to MCP servers, or those looking for actual MCP server implementations. ### [clintrials-mcp](https://mcpcatalogs.com/en/server/plainyogurt21-clintrials-mcp) by plainyogurt21 · score 33 · ★0 MCP server providing structured access to ClinicalTrials.gov data with search, retrieval, and analysis capabilities. **When to choose:** Choose this when you need structured access to clinical trial data for research or medical applications, particularly if you're already working with healthcare APIs or clinical research databases. **When NOT to choose:** Don't choose this if you need real-time clinical trial updates (this server depends on the ClinicalTrials.gov API schedule), or if you require highly specialized medical data beyond what ClinicalTrials.gov provides. ### [KoikatuMCP](https://mcpcatalogs.com/en/server/great-majority-koikatumcp) by great-majority · score 33 · ★4 MCP server that allows AI assistants to control Koikatu Studio through natural language commands. **When to choose:** Choose KoikatuMCP if you want to use AI to automate scene creation, character posing, or virtual photography in Koikatu Studio through natural language commands. **When NOT to choose:** Don't choose KoikatuMCP if you don't own or use Koikatu/Koikatsu Sunshine, or if you need integration with other 3D software beyond this specific game. ### [wisdomforge](https://mcpcatalogs.com/en/server/hadv-wisdomforge) by hadv · score 33 · ★4 Knowledge management MCP server using Qdrant vector database for storing and retrieving domain knowledge. **When to choose:** Choose WisdomForge when you need to integrate specialized knowledge bases into AI workflows, particularly for storing and retrieving domain-specific best practices and insights. **When NOT to choose:** Avoid if you need broader document management capabilities beyond specialized knowledge retrieval, or if you're looking for a solution with more comprehensive collaboration features. ### [markapatent-mcp](https://mcpcatalogs.com/en/server/andrianagnostic63-markapatent-mcp) by Andrianagnostic63 · score 33 · ★0 MCP server providing programmatic access to TÜRKPATENT's trademark, patent, and design databases through a standardized interface for IP research. **When to choose:** Choose this server when you need specialized access to Turkish intellectual property databases through an MCP interface, particularly for trademark searches and IP research workflows. **When NOT to choose:** Avoid this if you need access to patent databases outside of Turkey, as this server only provides access to TÜRKPATENT data. ### [langchain-mcp-server](https://mcpcatalogs.com/en/server/liteobject-langchain-mcp-server) by LiteObject · score 33 · ★1 Dual-mode MCP server providing live LangChain documentation, API references, and code examples from official sources. **When to choose:** Choose this server when you need real-time LangChain documentation and API references integrated into your AI workflow. **When NOT to choose:** Don't choose this if you need offline access to documentation or if you're working with a version of LangChain that's not yet reflected in the live sources. ### [a2a-go](https://mcpcatalogs.com/en/server/theapemachine-a2a-go) by TheApeMachine · score 33 · ★3 A2A-Go is a distributed AI agent framework implementing Google's A2A protocol with MCP interoperability. **When to choose:** Choose A2A-Go if you need a distributed agent architecture with Google's A2A protocol and want to integrate with MCP while maintaining Go's performance benefits. **When NOT to choose:** Avoid A2A-Go if you need a production-ready MCP server with complete implementation, as many MCP resource features are still marked as unfinished (incomplete resource management). ### [MCP-Clinic-Bot](https://mcpcatalogs.com/en/server/sankett13-mcp-clinic-bot) by sankett13 · score 33 · ★4 AI-powered medical chatbot using MCP with Google Gemini, ChromaDB vector search, and React frontend. **When to choose:** Choose this MCP server for healthcare providers who need an AI chatbot with appointment management capabilities and want to leverage MCP for integrating with other AI systems. **When NOT to choose:** Avoid this if you need healthcare compliance features beyond basic security, or if you're looking for a general-purpose chatbot framework that isn't specialized for medical contexts. ### [MCP-F1analisys](https://mcpcatalogs.com/en/server/maxbleu-mcp-f1analisys) by Maxbleu · score 32 · ★0 An MCP server providing Formula 1 racing analysis tools through the Model Context Protocol. **When to choose:** Choose this MCP server when you need detailed F1 racing analytics through an MCP interface, especially if you're working with Claude Desktop and want to integrate racing data into your workflow. **When NOT to choose:** Avoid this server if you need real-time F1 data, as it relies on FastF1 which may have delay in data availability after races, or if you need analytics beyond basic performance metrics. ### [MCP-F1data](https://mcpcatalogs.com/en/server/maxbleu-mcp-f1data) by Maxbleu · score 32 · ★0 A Python-based MCP server providing Formula 1 racing data through the F1DB API. **When to choose:** Choose this server when you need to integrate F1 racing data into AI applications or conversational interfaces with minimal setup. **When NOT to choose:** Avoid this if you need real-time F1 data as the server relies on F1DB which may have updated schedules, or if you require data outside historical F1 statistics. ### [mcp-prompts-catalog](https://mcpcatalogs.com/en/server/sparesparrow-mcp-prompts-catalog) by sparesparrow · score 32 · ★2 A curated catalog of prompts for MCP ecosystem with JSON/YAML/TXT/MD support. **When to choose:** Choose this MCP server if your team needs a centralized, version-controlled prompt catalog that supports multiple formats and simplifies prompt management across applications. **When NOT to choose:** Not suitable if you need custom authentication or extensive prompt transformation capabilities, as this server focuses primarily on storage and retrieval of basic prompt formats. ### [PopUI](https://mcpcatalogs.com/en/server/kelnishi-popui) by kelnishi · score 32 · ★8 PopUI enables collaborative UX in Claude Desktop with bidirectional UI tools. **When to choose:** Choose PopUI when you need to create interactive visual tools for collaborative work with Claude, especially for control interfaces, design tools, or games where real-time visual feedback matters. **When NOT to choose:** Don't choose PopUI if you're not on macOS, need cross-platform compatibility, or require integration with LLM services other than Claude Desktop. ### [my-monkey-app](https://mcpcatalogs.com/en/server/kardelruveyda-my-monkey-app) by KardelRuveyda · score 31 · ★1 Interactive console app displaying monkey information with MCP server integration. **When to choose:** Choose this for educational purposes or as a simple demonstration of MCP server integration with console applications. **When NOT to choose:** Not suitable for production use due to limited functionality and educational focus; lacks advanced features needed for serious applications. ### [mcp-hands-on-with-agentic-ai](https://mcpcatalogs.com/en/server/manulthanura-mcp-hands-on-with-agentic-ai) by manulthanura · score 31 · ★1 Educational repository with MCP server examples and templates for building agentic AI capabilities. **When to choose:** Choose this repository if you're learning MCP implementation or need examples for building MCP servers with specific functionality like text processing, weather data, or GitHub model comparison. **When NOT to choose:** Don't choose this for production-ready MCP servers or if you need extensive documentation and long-term support beyond educational examples. ### [mcp-server](https://mcpcatalogs.com/en/server/biggy44-mcp-server) by biggy44 · score 31 · ★1 ASON MCP server for efficient JSON compression/decompression to optimize token usage. **When to choose:** Choose this application if you need a simple desktop tool for compressing and decompressing JSON data with ASON format and don't require MCP server integration. **When NOT to choose:** Don't choose this if you actually need a proper MCP server implementation for integration with Claude Desktop or other MCP clients. ### [student-services-mcp](https://mcpcatalogs.com/en/server/edyvision-student-services-mcp) by EdyVision · score 31 · ★1 A FastAPI-based MCP server for student services management with academic planning, financial aid, and dropout risk prediction. **When to choose:** Educational institutions needing a student services demo with analytics capabilities and planning tools for academic advisors. **When NOT to choose:** Production environments requiring real student data integration, as this is a demo with synthetic data; systems needing more comprehensive authentication beyond basic token auth. ### [plausible-mcp](https://mcpcatalogs.com/en/server/the-focus-ai-plausible-mcp) by The-Focus-AI · score 31 · ★1 An MCP server for natural interaction with Plausible Analytics data via LLM conversations. **When to choose:** Choose this MCP server if you already use Plausible Analytics and want natural-language access to your website metrics through LLM conversations. **When NOT to choose:** Don't choose this if you need analytics platforms other than Plausible, or if you require write access to your analytics data (this is read-only). ### [feyod-mcp](https://mcpcatalogs.com/en/server/jeroenvdmeer-feyod-mcp) by jeroenvdmeer · score 31 · ★1 MCP server for Feyenoord football data queries via natural language interface. **When to choose:** Choose Feyod MCP when you need to query Feyenoord football data through natural language, especially if you're already using LLM services like OpenAI or Google. **When NOT to choose:** Don't choose this if you need data for other football clubs or require write access to the database, as it's read-only and Feyenoord-specific. ### [usgs-quakes-mcp](https://mcpcatalogs.com/en/server/blake365-usgs-quakes-mcp) by blake365 · score 31 · ★1 An MCP server providing natural language access to USGS earthquake data through search and detail tools. **When to choose:** Choose this server when you need quick, natural language access to USGS earthquake data without complex setup or authentication. **When NOT to choose:** Don't choose this if you need real-time earthquake notifications, historical data beyond USGS coverage, or require authenticated API access to third-party systems. ### [halo-infinite-mcp](https://mcpcatalogs.com/en/server/dend-halo-infinite-mcp) by dend · score 31 · ★6 MCP server for Halo Infinite game data, accessing player stats, match history, and marketplace information. **When to choose:** Choose this if you're a Halo Infinite player who wants quick access to your game data through your preferred MCP client without launching the full game. **When NOT to choose:** Don't choose this if you're looking for official Microsoft support or need features beyond what's currently implemented, as it's an alpha-quality third-party tool. ### [WeatherSync-AI](https://mcpcatalogs.com/en/server/harshildoshi153-weathersync-ai) by HarshilDoshi153 · score 30 · ★0 A lightweight MCP server that fetches real-time weather data from OpenWeather.org based on city names. **When to choose:** Choose this server when you need to add basic weather functionality to an AI agent with minimal setup. **When NOT to choose:** Avoid this server if you need detailed historical weather data, forecasting capabilities, or extensive meteorological information. ### [Model-Context-Protocol](https://mcpcatalogs.com/en/server/abdulabin-model-context-protocol) by Abdulabin · score 30 · ★0 A Python MCP client implementing multi-server connectivity with LLM orchestration via Gemini API. **When to choose:** When you need to build applications that orchestrate multiple MCP servers and prefer a Python implementation with Gemini integration. **When NOT to choose:** When you're looking for MCP servers to connect to, not a client implementation. This project doesn't provide any tools or resources that other clients can connect to. ### [grimoire-mcp](https://mcpcatalogs.com/en/server/linell-grimoire-mcp) by Linell · score 30 · ★0 An MCP server providing D&D 5e spell information through search tools and detailed spell data. **When to choose:** Choose this server if you need quick access to D&D 5e spell information during gameplay or character creation, especially when using MCP-compatible AI assistants. **When NOT to choose:** Avoid this server if you need data beyond D&D 5e spells, or if you require a more extensive game data API that includes monsters, items, and other game elements. ### [mcp-sse-proxy](https://mcpcatalogs.com/en/server/getsimpletool-mcp-sse-proxy) by getsimpletool · score 30 · ★3 A Python proxy that bridges MCP STDIO communication with SSE protocol for web-based ClaudeAI clients. **When to choose:** Choose this when you need to connect web-based clients to MCP servers that only support STDIO communication. **When NOT to choose:** Avoid if you need direct MCP tool implementation rather than a proxy, or if you require a more robust web integration solution. ### [nps-explorer-mcp-server](https://mcpcatalogs.com/en/server/kyle-ski-nps-explorer-mcp-server) by Kyle-Ski · score 29 · ★2 An MCP server for U.S. national parks data, providing park info, trails, alerts, events, and weather forecasts via API integrations. **When to choose:** Choose this MCP server when you need specialized access to US national park information through LLM tool calls, especially if you're already using Cloudflare infrastructure. **When NOT to choose:** Don't choose this if you need information about parks outside the US, or if you prefer solutions with more extensive API integrations beyond what this server provides. ### [mcp-untappd-server-dotnet](https://mcpcatalogs.com/en/server/jtucker-mcp-untappd-server-dotnet) by jtucker · score 29 · ★9 Untappd MCP Server using Azure Functions in F# for beer data integration. **When to choose:** Choose this implementation if you're already using Azure Functions and want to integrate Untappd data with AI assistants through an MCP server. **When NOT to choose:** Don't choose this if you're not on Azure or if you prefer a simpler implementation without the complexity of Azure Functions. ### [MCP_Server](https://mcpcatalogs.com/en/server/thomasjanssen-tech-mcp-server) by ThomasJanssen-tech · score 28 · ★1 Python MCP server for querying Crunchbase company data through Claude Desktop. **When to choose:** Choose this MCP server if you need to analyze Crunchbase company data through Claude Desktop and are willing to handle your own data ingestion with Bright Data API. **When NOT to choose:** Don't choose this server if you need real-time data updates, want built-in authentication, or prefer solutions that don't require external API keys like Bright Data. ### [remote-mcp-server-authless-gif-search](https://mcpcatalogs.com/en/server/elizabethsiegle-remote-mcp-server-authless-gif-search) by elizabethsiegle · score 28 · ★1 A Cloudflare-based remote MCP server for authenticating GIF searches without requiring authentication. **When to choose:** Choose this if you need a simple, authentication-free MCP server prototype or want to learn how to deploy MCP on Cloudflare Workers. **When NOT to choose:** Don't choose this for production use as it lacks authentication, or if you need a more feature-rich GIF search solution. ### [mcp-mindmesh](https://mcpcatalogs.com/en/server/wheattoast11-mcp-mindmesh) by wheattoast11 · score 28 · ★1 An MCP server that orchestrates multiple Claude 3.7 Sonnet instances in a quantum-inspired swarm with field coherence optimization. **When to choose:** Choose this server when you need ensemble AI reasoning with field coherence across multiple specialized instances, particularly for complex multi-domain analysis tasks. **When NOT to choose:** Avoid if you need a lightweight solution or have concerns about vendor lock-in to Anthropic's API and VoyageAI's embedding service. ### [search_docs_mcp](https://mcpcatalogs.com/en/server/mostafa-ghaith-search-docs-mcp) by mostafa-ghaith · score 28 · ★1 MCP server for semantic search across AI library documentation using Serper API **When to choose:** Choose this server when you need up-to-date documentation search across multiple AI libraries and are already using MCP-compatible tools like Claude Desktop or Cursor. **When NOT to choose:** Don't choose this if you need documentation for non-AI libraries, if the Serper API's rate limits are insufficient for your needs, or if you prefer offline solutions. ### [MCP-GROQ](https://mcpcatalogs.com/en/server/rahulsamant37-mcp-groq) by rahulsamant37 · score 28 · ★1 MCP-GROQ provides a suite of AI tools for search, mathematics, and news retrieval using MCP framework. **When to choose:** Choose MCP-GROQ when you need lightweight search, math, or news tools integrated into an MCP-based system and already using Groq's LLM services. **When NOT to choose:** Avoid MCP-GROQ if you need enterprise-level reliability, higher performance alternatives, or tools outside its narrow scope of search, math, and news aggregation. ### [aws-geocode-mcp](https://mcpcatalogs.com/en/server/kanahiro-aws-geocode-mcp) by Kanahiro · score 28 · ★1 An MCP server for geocoding using Amazon Location Service that converts addresses to coordinates. **When to choose:** Choose this when you need geocoding capabilities and are already using AWS services or prefer Amazon Location Service over alternatives. **When NOT to choose:** Avoid this if you don't want to use AWS services or if you need geocoding capabilities outside the AWS ecosystem. ### [AzureMCP-demo](https://mcpcatalogs.com/en/server/abhijithganesh-azuremcp-demo) by AbhijithGanesh · score 27 · ★0 A dual-language (Python/TypeScript) tutorial implementation of an MCP server for Azure services with clear setup instructions. **When to choose:** Choose this if you're looking for a minimal starting point to implement your own Azure MCP server. **When NOT to choose:** Not suitable for production use as it doesn't include actual Azure service integrations. ### [mcp-fetch-weather](https://mcpcatalogs.com/en/server/isseisuar-mcp-fetch-weather) by IsseiSuar · score 27 · ★0 A minimalist MCP server providing weather fetching functionality using Open-Meteo API. **When to choose:** Choose this for learning MCP implementation or as a starting point when building simple weather-integrated AI applications. **When NOT to choose:** Don't choose for production use as it lacks advanced features like error handling, authentication, and comprehensive weather data options. ### [ArXiv-Analyst](https://mcpcatalogs.com/en/server/devjothish-arxiv-analyst) by devjothish · score 25 · ★1 MCP-powered research assistant that searches and analyzes ArXiv papers with linguistic insights. **When to choose:** When you need a basic ArXiv paper search and analysis system with a simple web interface. **When NOT to choose:** When you need production-ready tools with comprehensive documentation, as this appears to be a proof-of-concept with minimal implementation. --- ## browser-automation (50) ### [chrome-devtools-mcp](https://mcpcatalogs.com/en/server/chromedevtools-chrome-devtools-mcp) by ChromeDevTools · score 65 · ★39,757 Chrome DevTools MCP server lets AI coding agents control and inspect live Chrome browsers through the MCP protocol. **When to choose:** When you need your AI coding agent to have full control and debugging capabilities for Chrome browser instances during development and testing workflows. **When NOT to choose:** If you need browser automation for browsers other than Chrome, or if you're concerned about Google collecting usage statistics despite the ability to opt-out. ### [httprunner](https://mcpcatalogs.com/en/server/httprunner-httprunner) by httprunner · score 55 · ★4,277 HttpRunner is an all-in-one testing framework that includes MCP server functionality for UI automation. **When to choose:** Choose HttpRunner's MCP server if you need cross-platform UI automation with visual AI capabilities and want to combine it with API testing in a unified framework. **When NOT to choose:** Avoid if you need traditional element locators (XPath, CSS selectors) or require testing frameworks with more mature community support. ### [stealth-browser-mcp](https://mcpcatalogs.com/en/server/vibheksoni-stealth-browser-mcp) by vibheksoni · score 54 · ★643 Stealth Browser MCP provides undetectable browser automation with antibypass capabilities for MCP-compatible AI agents. **When to choose:** Choose this when you need to automate interactions with websites that have sophisticated anti-bot protections or require pixel-perfect UI cloning capabilities. **When NOT to choose:** Avoid this if you need simple, straightforward browser automation without anti-bot bypass requirements or if you prefer a more established tool like Playwright. ### [opentabs](https://mcpcatalogs.com/en/server/opentabs-dev-opentabs) by opentabs-dev · score 53 · ★579 Browser automation server that authenticates with web APIs through browser sessions, exposing 100+ plugins with ~2,000 tools. **When to choose:** Choose OpenTabs when you need AI to interact with web applications you already use, especially when those services don't provide APIs or when you want to avoid managing API keys. **When NOT to choose:** Don't choose OpenTabs if you need direct API access to services that already provide well-documented APIs, or if you prefer not to use browser extensions for security reasons. ### [flyto-core](https://mcpcatalogs.com/en/server/flytohub-flyto-core) by flytohub · score 51 · ★270 An MCP server for AI agent automation with 412 modules, execution tracing, replay functionality, and built-in recipes. **When to choose:** Choose flyto-core when you need comprehensive browser automation with full traceability and replay capabilities, especially when working with complex multi-step workflows involving browser interaction, API calls, and file operations. **When NOT to choose:** Avoid flyto-core if you need simple, one-off browser interactions without the overhead of a full automation engine, or if you require native mobile app automation (flyto-core focuses on web browsers). ### [glance](https://mcpcatalogs.com/en/server/debugbase-glance) by DebugBase · score 51 · ★149 AI-powered browser automation MCP server for Claude Code with 30 tools for testing, screenshots, and interaction. **When to choose:** Choose Glance when you need Claude to directly interact with web applications through browser automation for testing, documentation, or visual verification tasks. **When NOT to choose:** Don't choose Glance if you need to test non-HTTP-based applications or if you require more sophisticated cross-browser testing beyond the Chromium family. ### [mcp](https://mcpcatalogs.com/en/server/browsermcp-mcp) by BrowserMCP · score 51 · ★6,510 Browser MCP enables AI applications to control your browser locally for automation tasks. **When to choose:** Choose Browser MCP when you need to automate your existing browser session with AI, maintaining logged-in state and avoiding basic bot detection. **When NOT to choose:** Avoid Browser MCP if you need browser automation for headless environments without GUI, or if you require cross-browser support beyond Chrome. ### [desktop](https://mcpcatalogs.com/en/server/agentify-sh-desktop) by agentify-sh · score 50 · ★391 MCP server for controlling web AI sessions (ChatGPT, Claude, etc.) via parallel tabs and automation. **When to choose:** Choose this when you need multiple AI tools to control your existing web service subscriptions without separate accounts, especially when working with context-heavy tasks requiring file uploads or artifact downloads. **When NOT to choose:** Avoid if you need read/write access to AI service APIs directly, or if you prefer a cloud-based solution rather than local browser automation. ### [firefox-devtools-mcp](https://mcpcatalogs.com/en/server/mozilla-firefox-devtools-mcp) by mozilla · score 50 · ★162 Mozilla's MCP server for automating Firefox DevTools via WebDriver BiDi, enabling AI assistants to control and inspect browser functionality. **When to choose:** Choose this when you need AI-powered browser automation specifically for Firefox, especially when working with web scraping, testing, or interaction with websites. **When NOT to choose:** Don't choose this if you need browser automation for browsers other than Firefox, or if you cannot install Firefox locally on your system. ### [charlotte](https://mcpcatalogs.com/en/server/ticktockbent-charlotte) by TickTockBent · score 50 · ★142 Token-efficient browser MCP server that provides structured web access to AI agents. **When to choose:** Choose Charlotte when token efficiency is critical and you need structured web access with controlled detail levels for AI agents. **When NOT to choose:** Don't choose Charlotte if you need raw HTML access or low-level DOM manipulation beyond what its structured representations provide. ### [mcp-screenshot-website-fast](https://mcpcatalogs.com/en/server/just-every-mcp-screenshot-website-fast) by just-every · score 50 · ★106 MCP server for fast webpage screenshots optimized for AI vision workflows with automatic tiling. **When to choose:** Choose this server when you need to capture webpage screenshots for AI analysis, particularly when working with vision models that have optimal input sizes around 1.15 megapixels. **When NOT to choose:** Don't choose this if you need screenshots larger than 1072x1072 pixels, or if you require authentication/cookie-based access to websites as it doesn't handle session management. ### [vibetest-use](https://mcpcatalogs.com/en/server/browser-use-vibetest-use) by browser-use · score 49 · ★793 MCP server for automated QA testing using Browser-Use agents to find UI bugs and accessibility issues. **When to choose:** Choose Vibetest when you need automated UI testing with multiple concurrent agents and want to integrate testing directly into your Claude or Cursor workflow through MCP. **When NOT to choose:** Avoid Vibetest if you need testing capabilities without Google API dependencies or if your project doesn't already use Browser-Use. ### [browser-use-mcp-server](https://mcpcatalogs.com/en/server/kontext-security-browser-use-mcp-server) by kontext-security · score 49 · ★822 MCP server enabling AI agents to control web browsers with browser-use. **When to choose:** Choose this MCP server when you need AI-powered browser automation for web scraping, testing, or browsing assistance with real-time VNC monitoring capabilities. **When NOT to choose:** Don't choose this if you need to support browser engines other than Chromium, require auth models for multi-user environments, or prefer simpler browser automation without AI components. ### [safari-mcp](https://mcpcatalogs.com/en/server/achiya-automation-safari-mcp) by achiya-automation · score 49 · ★63 Native Safari browser automation for AI agents with 80 tools via AppleScript, zero overhead, and lower CPU usage than Chrome DevTools. **When to choose:** Choose Safari MCP when you need to automate browser interactions on macOS while maintaining logged-in sessions and minimizing CPU usage. **When NOT to choose:** Don't choose Safari MCP if you need cross-platform support (Windows/Linux) or require Chrome DevTools Protocol-specific functionality. ### [byob](https://mcpcatalogs.com/en/server/wxtsky-byob) by wxtsky · score 49 · ★121 byob is a local MCP server that lets AI assistants control your existing Chrome browser with all your logged-in sessions. **When to choose:** When you need AI agents to interact with websites as a logged-in human would, without copying cookies or dealing with bot detection. **When NOT to choose:** If you need to automate headless browser interactions without a visible browser instance, or if you're working with websites that block automation frameworks. ### [web-agent-protocol](https://mcpcatalogs.com/en/server/ota-tech-ai-web-agent-protocol) by OTA-Tech-AI · score 48 · ★497 Web Agent Protocol (WAP) records browser interactions and converts them into MCP servers for web automation replay. **When to choose:** Choose this when you need to automate complex browser workflows and want to convert them into reusable MCP servers for AI agents. **When NOT to choose:** Avoid if you need non-browser automation or require real-time browser interaction without recording first. ### [hanzi-browse](https://mcpcatalogs.com/en/server/hanzili-hanzi-browse) by hanzili · score 48 · ★161 MCP server that enables AI agents to browse local websites with site-specific playbooks for complex interactions. **When to choose:** Choose Hanzi Browse when you need to automate browser interactions for AI agents on complex websites with dynamic content and anti-bot measures that standard scrapers can't handle. **When NOT to choose:** Don't choose Hanzi Browse if you need to automate simple sites that standard web scrapers can handle, or if you require open-source licensing (the core components are proprietary). ### [mcp-server](https://mcpcatalogs.com/en/server/browserstack-mcp-server) by browserstack · score 48 · ★139 BrowserStack's MCP server for comprehensive testing across real devices, browsers, and AI-powered test management. **When to choose:** Choose this server if you're already using BrowserStack for testing and want to integrate its capabilities directly into your AI-powered development workflow. **When NOT to choose:** Don't choose this if you need testing solutions that don't require BrowserStack's infrastructure, or if you're concerned about vendor lock-in to a specific testing platform. ### [camofox-mcp](https://mcpcatalogs.com/en/server/redf0x1-camofox-mcp) by redf0x1 · score 47 · ★61 Anti-detection browser MCP server with 47 automation tools for web interactions without blocking. **When to choose:** Choose CamoFox MCP when you need anti-detection browser automation for AI agents, particularly for login-heavy workflows or sites with strict bot detection mechanisms. **When NOT to choose:** Don't choose this if you need simple browser automation without anti-detection capabilities or if you want an all-in-one solution without managing separate browser server components. ### [parallel-browser-mcp](https://mcpcatalogs.com/en/server/etairl-parallel-browser-mcp) by etairl · score 47 · ★99 An MCP server enabling parallel browser automation across multiple cloud providers with session-based control. **When to choose:** Choose this when you need parallel browser automation across multiple cloud providers and want a consistent MCP interface regardless of the underlying browser technology. **When NOT to choose:** Avoid if you only need a single browser instance or if you're concerned about vendor lock-in with specific cloud providers. ### [open-claude-in-chrome](https://mcpcatalogs.com/en/server/noemica-io-open-claude-in-chrome) by noemica-io · score 47 · ★89 Open-source Chrome extension that enables unrestricted browser automation with Claude Code through MCP protocol. **When to choose:** Choose this when you need to access blocked websites with Claude's browser automation or want to avoid Anthropic's domain restrictions while maintaining identical tool functionality. **When NOT to choose:** Don't choose this if you prefer official Anthropic support or need features beyond what's available in the original Claude extension. ### [puppeteer-mcp-claude](https://mcpcatalogs.com/en/server/jaenster-puppeteer-mcp-claude) by jaenster · score 47 · ★38 Browser automation MCP server for Claude using Puppeteer to control web pages, capture screenshots, and interact with websites. **When to choose:** Choose this when you need browser automation within Claude, especially for web scraping, form filling, or testing dynamic JavaScript-driven websites. **When NOT to choose:** Not ideal for production systems requiring high reliability or non-Chromium browsers; opt for dedicated browser automation frameworks instead. ### [comet-mcp](https://mcpcatalogs.com/en/server/hanzili-comet-mcp) by hanzili · score 47 · ★155 MCP server connecting Claude Code to Perplexity Comet browser for intelligent web browsing and research. **When to choose:** Choose Comet MCP when you need complex web browsing capabilities beyond simple search APIs, especially for sites requiring login, navigation, or handling dynamic content. **When NOT to choose:** Avoid if you need direct browser control without the Comet dependency, or if you're working on platforms not supported (macOS, Windows, WSL2 only). ### [DrissionPageMCP](https://mcpcatalogs.com/en/server/wxhzhwxhzh-drissionpagemcp) by wxhzhwxhzh · score 46 · ★214 A browser automation MCP server using DrissionPage and FastMCP for AI-controlled web operations. **When to choose:** Choose DrissionPageMCP when you need browser automation capabilities through MCP for AI agents, especially when working with Chinese web pages that require robust element handling. **When NOT to choose:** Avoid if you need cross-browser support beyond Chrome, or if you're working in a restricted environment where installing additional browsers isn't feasible. ### [flowlens-mcp-server](https://mcpcatalogs.com/en/server/magentic-flowlens-mcp-server) by magentic · score 46 · ★105 FlowLens MCP server provides coding agents with complete browser context for debugging and regression testing. **When to choose:** Choose FlowLens when you need to debug browser applications or create regression tests and want AI agents to have complete context without reproducing issues. **When NOT to choose:** Avoid if you don't use Chrome-based browsers or if you need to record activities outside the browser environment. ### [perplexity-mcp-zerver](https://mcpcatalogs.com/en/server/wysh3-perplexity-mcp-zerver) by wysh3 · score 46 · ★93 Minimalist MCP server for AI-powered web research using Perplexity without API keys through browser automation. **When to choose:** Choose this when you need web search and research capabilities without API key limitations or when you want persistent research conversations stored locally. **When NOT to choose:** Don't choose this if you need highly stable/reliable access as it depends on Perplexity's website consistency, or if you require authenticated API access to Perplexity's backend. ### [vessel-browser](https://mcpcatalogs.com/en/server/unmodeled-tyler-vessel-browser) by unmodeled-tyler · score 46 · ★76 Chromium-based browser built for autonomous agents with MCP control and durable state persistence. **When to choose:** Choose Vessel when building autonomous agents that require persistent browser state and need humans to supervise their web activities through a visible interface. **When NOT to choose:** Avoid Vessel if you need headless browser automation, require write access to browser data, or need a browser with security assurances. ### [open-browser-use](https://mcpcatalogs.com/en/server/ifuryst-open-browser-use) by iFurySt · score 46 · ★83 Platform-neutral browser automation MCP server for AI agents with CLI and SDK support. **When to choose:** Choose this for Chrome-based browser automation in AI agents when you need an open-source solution with SDK support in multiple languages. **When NOT to choose:** Don't choose if you need browser automation for non-Chrome browsers or require more advanced browser control beyond basic tab and element interaction. ### [cortex-scout](https://mcpcatalogs.com/en/server/cortex-works-cortex-scout) by cortex-works · score 46 · ★65 Unified web extraction and stateful automation engine for AI agents with MCP integration. **When to choose:** Choose CortexScout when your AI agents need reliable web scraping and stateful browser automation with anti-bot protection, especially when working with complex authentication flows or JavaScript-heavy sites. **When NOT to choose:** Avoid if you need open-source licensing (CortexScout uses MIT license but is part of a commercial ecosystem), or if your use case requires only simple static page scraping without browser automation. ### [webclaw](https://mcpcatalogs.com/en/server/kuroko1t-webclaw) by kuroko1t · score 45 · ★18 WebClaw is a Chrome extension-based MCP server that allows AI assistants to control real Chrome browser sessions through a WebSocket connection. **When to choose:** Choose WebClaw when you need to automate authenticated workflows in your real browser with existing sessions and when resistance to bot detection is important. **When NOT to choose:** Don't choose WebClaw if you need headless browser automation without a visible Chrome window or if you require cross-browser support beyond Chrome. ### [kernel-mcp-server](https://mcpcatalogs.com/en/server/kernel-kernel-mcp-server) by kernel · score 45 · ★30 Kernel MCP server provides secure cloud-browser automation and app management through MCP protocol. **When to choose:** Choose this MCP server when you need browser automation capabilities with AI assistants, especially if you're already using or want to use the Kernel platform for cloud-based browser sessions. **When NOT to choose:** Avoid this server if you need self-hosted browser automation without cloud dependency or if you prefer open-source tools without vendor lock-in to the Kernel platform. ### [MCPSafari](https://mcpcatalogs.com/en/server/epistates-mcpsafari) by Epistates · score 45 · ★23 Native Safari MCP server providing 23 tools for AI agents to control Safari on macOS. **When to choose:** Choose MCPSafari when you need Safari-specific automation on macOS, especially if you require native accessibility features and want to avoid Chrome's resource overhead. **When NOT to choose:** Don't choose MCPSafari if you need cross-platform browser automation or require Chrome-specific browser extensions and APIs. ### [mcp-browser-agent](https://mcpcatalogs.com/en/server/imprvhub-mcp-browser-agent) by imprvhub · score 45 · ★38 An MCP server providing Claude with autonomous browser automation capabilities including DOM manipulation, JavaScript execution, and API requests. **When to choose:** Choose this when you need Claude to autonomously interact with web applications, perform data extraction, or automate browser-based workflows. **When NOT to choose:** Avoid if you need browser automation outside of Claude Desktop, or if you require headless browser operations without visual debugging capabilities. ### [sample-agentic-ai-web](https://mcpcatalogs.com/en/server/aws-samples-sample-agentic-ai-web) by aws-samples · score 45 · ★50 Amazon Bedrock web automation assistant with MCP server for browser automation and vision capabilities. **When to choose:** Choose this MCP server if you need a production-ready web automation solution with vision capabilities and are already using AWS Bedrock. **When NOT to choose:** Don't choose this if you need vendor-agnostic solutions or aren't using AWS Bedrock, as it has tight AWS integration. ### [opencode-browser](https://mcpcatalogs.com/en/server/michaljach-opencode-browser) by michaljach · score 45 · ★46 MCP server plugin for OpenCode that enables browser automation through Browser MCP integration. **When to choose:** Choose this when you need AI-driven browser automation specifically within OpenCode editor, with optimized performance for web interactions. **When NOT to choose:** Don't choose if you need automation outside of OpenCode, require support for browsers beyond Chrome/Edge, or prefer a standalone browser automation tool. ### [plasmate](https://mcpcatalogs.com/en/server/plasmate-labs-plasmate) by plasmate-labs · score 45 · ★21 Plasmate is a browser engine for agents that converts HTML to a compact Semantic Object Model (SOM) via MCP. **When to choose:** Choose Plasmate when you need to process HTML content for AI agents with significant token compression and JavaScript execution capabilities. **When NOT to choose:** Don't choose Plasmate if you need to handle extremely large-scale concurrent sessions (500+), as current capacity is limited. ### [WEBGhosting-MCP](https://mcpcatalogs.com/en/server/yranjan06-webghosting-mcp) by yranjan06 · score 44 · ★29 Intelligent stealth browser MCP server with anti-fingerprinting for AI agents to browse the web undetected. **When to choose:** Choose WEBGhosting when you need AI agents to browse the web stealthily with human-like behavior and bypass anti-bot measures. **When NOT to choose:** Avoid this if you need headless browser automation without the stealth features, or if you prefer Python-based MCP servers. ### [cdpilot](https://mcpcatalogs.com/en/server/mehmetnadir-cdpilot) by mehmetnadir · score 44 · ★25 Zero-dependency browser automation CLI with MCP server for AI agents, offering 70+ commands and stealth mode. **When to choose:** Choose cdpilot when you need lightweight, zero-dependency browser automation for AI agents, especially when stealth features and minimal token usage are priorities. **When NOT to choose:** Don't choose cdpilot if you need complex browser interactions beyond its 70+ commands, or if you require support for browsers not in the Chrome family (Firefox, Safari). ### [webmcp-bridge](https://mcpcatalogs.com/en/server/holon-run-webmcp-bridge) by holon-run · score 44 · ★25 A bridge that connects local MCP clients to browser WebMCP tools through Playwright, supporting both native and injected adapters. **When to choose:** Choose webmcp-bridge when you need MCP access to browser-only services or when you want human + AI collaboration through the same authenticated browser session. **When NOT to choose:** Don't use this if you need persistent credential management or if your target sites block automated access through browser automation tools. ### [real-browser-mcp](https://mcpcatalogs.com/en/server/ofershap-real-browser-mcp) by ofershap · score 43 · ★20 MCP server + Chrome extension enabling AI agents to control real browser sessions with existing logins and cookies. **When to choose:** Choose this when your AI agent needs to interact with your authenticated browser sessions without re-authentication or when you need to maintain state between AI actions. **When NOT to choose:** Don't choose this for headless browser automation scenarios where you don't need a visual browser or if you're looking for a cross-browser solution beyond Chrome. ### [puppeteer-mcp-server](https://mcpcatalogs.com/en/server/sultannaufal-puppeteer-mcp-server) by sultannaufal · score 42 · ★19 Self-hosted Puppeteer MCP server with browser automation tools, supporting multiple transports and API key authentication. **When to choose:** Choose this server when you need comprehensive browser automation through MCP, especially for scenarios requiring advanced mouse interactions or cookie management for authentication. **When NOT to choose:** Avoid if you need headless-only operations (this provides full browser control) or if you require a managed cloud service rather than self-hosted solution. ### [mcp-aoai-web-browsing](https://mcpcatalogs.com/en/server/kimtth-mcp-aoai-web-browsing) by kimtth · score 41 · ★33 MCP server with Playwright browser automation via FastMCP, supporting both Azure and OpenAI APIs. **When to choose:** Choose this when you need MCP-based browser automation with OpenAI integration and prefer using Playwright for web control. **When NOT to choose:** Avoid if you need extensive browser features beyond navigation or are not already using Azure/OpenAI services. ### [izan.io](https://mcpcatalogs.com/en/server/ekingunoncu-izan-io) by ekingunoncu · score 41 · ★29 Chrome extension that turns any browser into an MCP server for AI automation. **When to choose:** Choose izan.io when you need AI agents to interact with web applications that lack APIs, or for complex browser automation tasks requiring JavaScript control. **When NOT to choose:** Avoid if you need browser automation outside Chrome/Chromium, or if you require a solution with official enterprise support and a more permissive license than AGPL-3.0. ### [OmniMCP](https://mcpcatalogs.com/en/server/openadaptai-omnimcp) by OpenAdaptAI · score 37 · ★71 OmniMCP enables AI models to understand and interact with UI through visual perception and precise control using MCP. **When to choose:** Choose OmniMCP when you need AI agents to interact with desktop applications through visual UI understanding and automated actions. **When NOT to choose:** Don't choose OmniMCP if you need web automation (it's focused on desktop UIs), if you're on Windows, or if you require production-ready stability. ### [mcp-link](https://mcpcatalogs.com/en/server/aurafriday-mcp-link) by AuraFriday · score 37 · ★24 Browser extension + MCP server that enables AI agents to use controlled local/remote tools. **When to choose:** Choose MCP Link when you want AI agents to access real-world tools through a browser interface with granular user control, especially when needing both remote and local tool access with Docker sandboxing capabilities. **When NOT to choose:** Avoid MCP Link if you need an open-source license (the source is visible but not forkable) or if you're looking for a solution without proprietary components or the 'visible source' restriction model. ### [rec-us-mcp-server](https://mcpcatalogs.com/en/server/elizabethsiegle-rec-us-mcp-server) by elizabethsiegle · score 37 · ★10 MCP server for San Francisco tennis court booking with authentication via browser automation. **When to choose:** Choose this if you're a regular San Francisco tennis player who wants to automate court bookings through AI assistants. **When NOT to choose:** Don't choose this if you don't live in San Francisco or don't play tennis regularly, as it's highly specific to this single use case. ### [autoapply-mcp](https://mcpcatalogs.com/en/server/preetrajdeo-autoapply-mcp) by preetrajdeo · score 36 · ★0 An MCP server that automates job applications by filling forms and uploading resumes via browser automation. **When to choose:** Choose AutoApply MCP when you're applying to multiple jobs with standard application forms and want to automate the repetitive filling of personal information, resume uploads, and basic qualifications. **When NOT to choose:** Don't choose this if you need custom application workflows, have concerns about server-side storage of personal data, or need to handle highly specialized application portals not supported by its browser automation. ### [MCP](https://mcpcatalogs.com/en/server/selviakdag-mcp) by SelviAkdag · score 33 · ★0 Windows-specific MCP server with browser automation and file management tools that actually work without Docker or WSL. **When to choose:** Choose this if you're working on Windows and need MCP agents for browser automation or file management without virtualization overhead. **When NOT to choose:** Don't choose this if you're on Linux/Mac, need advanced features not covered by the provided agents, or prefer a more general MCP solution. ### [mcpX](https://mcpcatalogs.com/en/server/rudra78996-mcpx) by Rudra78996 · score 31 · ★1 mcpX bridges AI applications with browser automation through MCP, enabling sophisticated web interaction workflows. **When to choose:** Choose mcpX when you need precise browser automation capabilities integrated with AI agents, particularly for testing, data collection, or complex web workflows requiring visual verification. **When NOT to choose:** Avoid mcpX if you need support for non-Chromium browsers or require advanced browser features not yet implemented, as it's still a relatively new project with limited browser compatibility. ### [browser-mcp-lite](https://mcpcatalogs.com/en/server/karolaapiarian589-browser-mcp-lite) by karolaapiarian589 · score 30 · ★0 A lightweight MCP server for Chrome browser automation through extensions with secure token authentication. **When to choose:** Choose this for lightweight browser automation when you need secure access through token auth and a minimal setup without a large tool stack. **When NOT to choose:** Don't choose this if you need cross-browser support, extensive automation features, or plan to use it in non-Windows environments. --- ## cloud-storage (5) ### [mcp](https://mcpcatalogs.com/en/server/awslabs-mcp) by awslabs · score 61 · ★9,068 AWS Labs provides a comprehensive suite of MCP servers for AWS services, enabling AI applications to access AWS documentation and resources. **When to choose:** Teams using AWS services that need AI assistants to access up-to-date AWS documentation, perform infrastructure operations, and follow AWS best practices without exposing credentials. **When NOT to choose:** If you need write access to AWS resources (these primarily provide read-only access), or if you prefer a more integrated solution like the newer Agent Toolkit for AWS with additional security features. ### [Fabric-Analytics-MCP](https://mcpcatalogs.com/en/server/santhoshravindran7-fabric-analytics-mcp) by santhoshravindran7 · score 46 · ★111 MCP server enabling AI assistants to securely access and analyze Microsoft Fabric Analytics data through authenticated API calls. **When to choose:** Choose this MCP server if you're already using Microsoft Fabric for analytics and want AI-powered workspace management, Spark monitoring, and automated migration tools. **When NOT to choose:** Don't choose this if you're not using Microsoft Fabric, as it requires authentication with Fabric services and provides no value for other data platforms. ### [google-cloud-mcp](https://mcpcatalogs.com/en/server/krzko-google-cloud-mcp) by krzko · score 45 · ★78 A comprehensive MCP server providing access to multiple Google Cloud services including billing, monitoring, IAM, Spanner, and more. **When to choose:** Choose this MCP server if you're already using Google Cloud services and need AI assistance with cost management, debugging, or infrastructure monitoring without switching between tools. **When NOT to choose:** Avoid this server if you're not using Google Cloud services or prefer vendor-agnostic solutions, as it's tightly coupled to Google Cloud and may introduce single-vendor lock-in concerns. ### [aws-s3-mcp](https://mcpcatalogs.com/en/server/samuraikun-aws-s3-mcp) by samuraikun · score 42 · ★24 A well-designed MCP server for AWS S3 integration with both HTTP and STDIO transport support. **When to choose:** Choose this server when you need to integrate S3 access into LLM applications, especially when working with Claude Desktop or requiring both STDIO and HTTP transport options. **When NOT to choose:** Don't choose this if you need write access to S3 (it's read-only) or if you need fine-grained access control beyond simple bucket filtering. ### [mcp-ipfs](https://mcpcatalogs.com/en/server/alexbakers-mcp-ipfs) by alexbakers · score 35 · ★16 An MCP server that wraps the w3 CLI to enable language models to interact with IPFS via storacha.network. **When to choose:** Choose this MCP server when working with IPFS through storacha.network and need AI integration for decentralized content management, especially if you're already using the w3 CLI. **When NOT to choose:** Avoid this if you need direct IPFS access without the storacha.network abstraction, or if you require filesystem path handling in containerized environments which can be complex. --- ## communication (65) ### [slack-mcp-server](https://mcpcatalogs.com/en/server/korotovsky-slack-mcp-server) by korotovsky · score 56 · ★1,603 A feature-rich MCP server for Slack with stealth mode, OAuth support, and comprehensive message management tools. **When to choose:** Choose this server if you need comprehensive Slack integration with an MCP server, especially if you want stealth mode capabilities or need to avoid traditional bot installations. **When NOT to choose:** Don't choose this if you need write access to Slack by default (posting is disabled for safety), or if you're using bot tokens (xoxb-) which limit search functionality. ### [voicemode](https://mcpcatalogs.com/en/server/mbailey-voicemode) by mbailey · score 55 · ★1,181 VoiceMode enables natural voice conversations with Claude Code and other MCP agents through local or cloud speech services. **When to choose:** Choose VoiceMode when you need hands-free or eyes-free interaction with Claude Code, particularly during physical activities when typing is impractical. **When NOT to choose:** Don't choose VoiceMode if you're in a quiet environment where speaking would be disruptive, or if you need text-based conversation history. ### [slackdump](https://mcpcatalogs.com/en/server/rusq-slackdump) by rusq · score 55 · ★2,587 Slackdump MCP server provides tools to analyze Slack archives without admin privileges. **When to choose:** Choose this when you need to analyze Slack archives without admin privileges and want to use AI tools for content analysis. **When NOT to choose:** Don't choose this if you need real-time access to live Slack data or require write permissions to modify Slack content. ### [agent-kit](https://mcpcatalogs.com/en/server/keyid-ai-agent-kit) by KeyID-AI · score 53 · ★689 MCP email server with 27 tools for inbox, sending, contacts and auto-reply functionality. **When to choose:** Choose this server when you need email capabilities for AI agents without the complexity of setting up traditional email accounts or managing API keys. **When NOT to choose:** Not suitable if you need to connect to existing email accounts (Gmail, Outlook) or require advanced email filtering rules beyond what the MCP server provides. ### [mac_messages_mcp](https://mcpcatalogs.com/en/server/carterlasalle-mac-messages-mcp) by carterlasalle · score 52 · ★284 A Python MCP server that bridges macOS Messages app with LLMs for secure querying and analysis of iMessage conversations. **When to choose:** Choose this MCP server when you need to integrate macOS Messages functionality into AI workflows, especially if you work with a mix of iMessage and Android contacts. **When NOT to choose:** Don't choose this if you need cross-platform message management beyond macOS, as it's limited to Apple's Messages app database. ### [discord-mcp](https://mcpcatalogs.com/en/server/saseq-discord-mcp) by SaseQ · score 51 · ★310 A Java-based MCP server that enables AI assistants to interact with Discord via comprehensive channel, message, and user management tools. **When to choose:** Choose this server if you need comprehensive Discord automation through AI assistants and want production-ready deployment options with Docker support. **When NOT to choose:** Don't choose this if you need a lightweight solution with minimal dependencies, as it requires Java and Docker for deployment. ### [line-bot-mcp-server](https://mcpcatalogs.com/en/server/line-line-bot-mcp-server) by line · score 51 · ★583 LINE Bot MCP Server connects AI agents to LINE Official Accounts via 12 tools for messaging, profile retrieval, and rich menu management. **When to choose:** Choose this when you need to connect AI agents with LINE's messaging platform for customer service, notifications, or user engagement through LINE Official Accounts. **When NOT to choose:** Avoid if you're not already using LINE Official Accounts or if you need messaging functionality outside of LINE's ecosystem. ### [telegram-mcp](https://mcpcatalogs.com/en/server/chaindead-telegram-mcp) by chaindead · score 50 · ★326 A Go-based MCP server that bridges Telegram API with AI assistants for managing messages, dialogs, and drafts. **When to choose:** Choose this server if you need to integrate Telegram functionality with AI assistants for message management and organization, especially if you're already using MCP-compatible clients like Claude Desktop or Cursor. **When NOT to choose:** Don't choose this server if you need extensive Telegram bot management features or if you have concerns about using third-party tools with your personal Telegram account due to API terms of service. ### [mcp-teams-server](https://mcpcatalogs.com/en/server/inditextech-mcp-teams-server) by InditexTech · score 50 · ★373 MCP server for Microsoft Teams integration with message handling, member mentions, and thread management. **When to choose:** Choose this MCP server if you need to integrate Microsoft Teams into AI workflows or create agents that can interact with Teams conversations programmatically. **When NOT to choose:** Avoid if you need comprehensive Teams management capabilities beyond messaging (like meeting scheduling or file management) or if you prefer a solution with less Azure dependency. ### [beever-atlas](https://mcpcatalogs.com/en/server/beever-ai-beever-atlas) by Beever-AI · score 50 · ★328 Beever Atlas is an MCP server that transforms team chats into a self-maintaining wiki with integrated search and QA capabilities. **When to choose:** Choose Beever Atlas if your team already uses Slack/Discord/Teams and wants to automatically transform conversations into a structured knowledge base accessible via MCP for AI agents. **When NOT to choose:** Avoid Beever Atlas if you need full control over data storage (it uses Weaviate, Neo4j, MongoDB, Redis), require a self-hosted solution without cloud dependencies, or need to process documents outside of team chat platforms. ### [ghost-mcp](https://mcpcatalogs.com/en/server/mfydev-ghost-mcp) by MFYDev · score 50 · ★186 An MCP server for managing Ghost CMS through LLM interfaces like Claude with comprehensive blog management tools. **When to choose:** Choose this server if you're already using Ghost CMS and want to manage your blog through natural language interfaces like Claude. **When NOT to choose:** Don't choose this if you're not using Ghost CMS, as it provides no value for other platforms. ### [fastmail-mcp](https://mcpcatalogs.com/en/server/madllama25-fastmail-mcp) by MadLlama25 · score 49 · ★119 A comprehensive MCP server providing access to Fastmail's email, contacts, and calendar APIs for AI assistants. **When to choose:** Choose this server if you need comprehensive AI access to Fastmail's email, contacts, and calendar services with rich functionality including threading, attachments, and bulk operations. **When NOT to choose:** Don't choose this if you use a different email provider as it's specifically built for Fastmail's API and authentication system. ### [teams-mcp](https://mcpcatalogs.com/en/server/floriscornel-teams-mcp) by floriscornel · score 49 · ★99 MCP server for Microsoft Teams and Graph API access with comprehensive messaging, search, and user management capabilities. **When to choose:** Choose this when you need comprehensive access to Microsoft Teams within an AI assistant, including messaging, user management, and file operations with proper authentication. **When NOT to choose:** Avoid if you need non-Microsoft chat platform integration or if your organization doesn't use Microsoft 365/Teams. ### [apple-mail-mcp](https://mcpcatalogs.com/en/server/patrickfreyer-apple-mail-mcp) by patrickfreyer · score 49 · ★119 A comprehensive MCP server providing full access to Apple Mail with 22 tools for reading, searching, composing, organizing, and analyzing emails. **When to choose:** Choose this server if you're a macOS user who wants AI-powered management of Apple Mail through natural language commands, especially if you're already using Claude Code or Claude Desktop. **When NOT to choose:** Avoid this if you don't use Apple Mail on macOS, need cross-platform email management, or require integration with non-Apple email clients or services. ### [WeChat-MCP](https://mcpcatalogs.com/en/server/biboyqg-wechat-mcp) by BiboyQG · score 48 · ★191 An MCP server that automates WeChat on macOS, allowing LLMs to read messages and send replies. **When to choose:** Choose this MCP server if you need AI integration with WeChat on macOS and want specialized automation features like message fetching, auto-replies, and Moments posting. **When NOT to choose:** Don't choose this if you're not using macOS, need Windows/Linux support, or require broader messaging platform support beyond WeChat. ### [mcp-discord](https://mcpcatalogs.com/en/server/barryyip0625-mcp-discord) by barryyip0625 · score 48 · ★84 A complete MCP server for Discord integration, enabling AI assistants to manage servers, channels, messages, roles, and more. **When to choose:** Choose this MCP server if you need to integrate AI capabilities with Discord for community management, automation, or building intelligent Discord bots. **When NOT to choose:** Don't choose this if you need access to Discord's voice features, real-time events, or if you require permissions beyond what Discord's API allows. ### [email-mcp](https://mcpcatalogs.com/en/server/codefuturist-email-mcp) by codefuturist · score 47 · ★45 Full-featured MCP email server with IMAP/SMTP support, OAuth2, and AI-powered triage across multiple accounts. **When to choose:** Choose this server when you need comprehensive email management with multi-account support, advanced features like email scheduling and AI triage, and integration with multiple MCP clients. **When NOT to choose:** Avoid if you only need basic email reading/sending functionality or prefer OAuth2 over experimental token authentication. ### [claude-telegram-supercharged](https://mcpcatalogs.com/en/server/k1p1l0-claude-telegram-supercharged) by k1p1l0 · score 47 · ★101 Enhanced Telegram plugin for Claude Code with threading, voice messages, stickers, GIFs, and more. **When to choose:** Users looking for enhanced Telegram integration with Claude Code who need features like voice messages, threaded conversations, and media support that aren't available in the official plugin. **When NOT to choose:** You shouldn't use this if you're looking for a true MCP server implementation or want a plugin that automatically updates without manual intervention after official plugin releases. ### [help-scout-mcp-server](https://mcpcatalogs.com/en/server/drewburchfield-help-scout-mcp-server) by drewburchfield · score 47 · ★40 MCP server for Help Scout - search conversations, threads, and inboxes with AI agents. **When to choose:** Choose this server if you're already using Help Scout for customer support and want AI agents to directly search and analyze conversations without switching contexts. **When NOT to choose:** Don't choose this if you need write access to Help Scout data (this is read-only only) or if you require compliance standards that conflict with message redaction options. ### [agenticmail](https://mcpcatalogs.com/en/server/agenticmail-agenticmail) by agenticmail · score 47 · ★121 AgenticMail provides AI agents with email/SMS infrastructure, enabling real communication via programmatic access. **When to choose:** Choose AgenticMail when building AI agents that need to communicate with the real world via email/SMS, especially when requiring isolated communication channels and security guardrails against sensitive data leakage. **When NOT to choose:** Avoid if you need a cloud-based solution without self-hosting requirements, or if you're looking for a simpler email integration without the agent-specific features. ### [zendesk-mcp-server](https://mcpcatalogs.com/en/server/reminia-zendesk-mcp-server) by reminia · score 47 · ★96 A comprehensive MCP server for Zendesk integration with ticket management and knowledge base access. **When to choose:** Choose this MCP server if you're using Zendesk for customer support and want to integrate AI assistance directly into your ticket management workflow without building custom integrations. **When NOT to choose:** Don't choose this if you need to integrate with a different help desk system or require functionality beyond what Zendesk's API provides. ### [mcp-telegram](https://mcpcatalogs.com/en/server/dryeab-mcp-telegram) by dryeab · score 46 · ★243 MCP server enabling LLMs to interact with Telegram through messaging, search, and media operations. **When to choose:** Choose this when you need to integrate Telegram capabilities into AI workflows or create AI-powered Telegram bots with comprehensive message handling features. **When NOT to choose:** Avoid if you need to handle sensitive communications where privacy is paramount, as Telegram's MTProto protocol has had security scrutiny in the past. ### [ntfy-me-mcp](https://mcpcatalogs.com/en/server/gitmotion-ntfy-me-mcp) by gitmotion · score 46 · ★59 A MCP server for AI agents to send and fetch notifications via ntfy service with token authentication. **When to choose:** Choose this MCP server when you need AI agents to send notifications to your devices via ntfy, especially if you're already using ntfy or want simple notification capabilities without complex infrastructure. **When NOT to choose:** Don't choose this if you need more robust notification features like email delivery, rich media attachments, or if you're looking for an enterprise-grade notification system with advanced routing and policies. ### [imessage-query-fastmcp-mcp-server](https://mcpcatalogs.com/en/server/hannesrudolph-imessage-query-fastmcp-mcp-server) by hannesrudolph · score 45 · ★77 FastMCP server providing safe access to iMessage database with proper validation and attachment handling. **When to choose:** Choose this if you need to analyze iMessage conversations on macOS and want a secure, read-only interface that properly handles phone numbers and attachments. **When NOT to choose:** Don't use this if you need write access to messages or if you're not on macOS, as the server specifically requires macOS for iMessage database access. ### [golemcore-bot](https://mcpcatalogs.com/en/server/alexk-dev-golemcore-bot) by alexk-dev · score 45 · ★40 GolemCore Bot is an agent platform with MCP integration, enabling channel-connected agents with plugins and memory. **When to choose:** Choose GolemCore Bot if you need a production-ready agent platform with built-in MCP integration for enterprise-scale AI applications requiring long-running workflows and memory management. **When NOT to choose:** Don't choose GolemCore Bot if you need a lightweight, simple solution or if you're not comfortable with Java-based deployments and Docker requirements. ### [whatsapp-mcp](https://mcpcatalogs.com/en/server/verygoodplugins-whatsapp-mcp) by verygoodplugins · score 45 · ★49 WhatsApp MCP server enables Claude to read and send WhatsApp messages with full media support. **When to choose:** Choose this server when you need to integrate WhatsApp functionality with Claude for automated messaging, archiving, or AI-powered communication workflows. **When NOT to choose:** Avoid this if you need end-to-end encryption assurance beyond local storage, or if you're looking for an enterprise-grade solution with multiple-user support. ### [agent-inbox](https://mcpcatalogs.com/en/server/gsd-build-agent-inbox) by gsd-build · score 45 · ★42 MCP server providing temporary email addresses for AI agents to handle email verification and auth flows automatically. **When to choose:** Choose agent-inbox when you need AI agents to handle email verification flows without API keys or manual intervention. **When NOT to choose:** Don't choose this if you need persistent email addresses or if services block disposable email domains, as inboxes don't survive server restarts. ### [whatsapp-mcp](https://mcpcatalogs.com/en/server/felipeadeildo-whatsapp-mcp) by felipeadeildo · score 45 · ★70 WhatsApp MCP server integrates WhatsApp conversations with AI assistants through MCP tools, enabling message reading, searching, and sending. **When to choose:** Choose this server when you need AI assistance to interact with your WhatsApp conversations, especially for summarizing, searching, or automating responses across multiple chats. **When NOT to choose:** Avoid this if you need enterprise-level WhatsApp integration, official API support, or if you're concerned about using unofficial WhatsApp APIs that may violate terms of service. ### [instagram_dm_mcp](https://mcpcatalogs.com/en/server/trypeggy-instagram-dm-mcp) by trypeggy · score 45 · ★167 Instagram DM MCP server enabling direct messaging, media sharing, and user interaction through Instagram's API. **When to choose:** Choose this MCP server if you need to programmatically interact with Instagram DMs from an AI agent, especially for social media management, customer support automation, or content archiving purposes. **When NOT to choose:** Don't choose this if you need access to Instagram's API for commercial applications without reverse engineering, as this server bypasses official API and may violate Instagram's terms of service. ### [telegram-api-mcp](https://mcpcatalogs.com/en/server/timoncool-telegram-api-mcp) by timoncool · score 45 · ★20 Comprehensive Telegram Bot API MCP server with full v9.6 coverage, meta-mode, and advanced rate limiting. **When to choose:** Choose this server if you need comprehensive Telegram bot integration with Claude AI, especially if you're working with the latest Bot API features and require advanced features like meta-mode for token efficiency. **When NOT to choose:** Consider alternatives if you need a different messaging platform API or require features beyond what Telegram Bot API offers, as this is exclusively for Telegram. ### [claude-ipc-mcp](https://mcpcatalogs.com/en/server/jdez427-claude-ipc-mcp) by jdez427 · score 45 · ★132 MCP server enabling AI-to-AI communication across different platforms like Claude, Gemini, and ChatGPT using natural language commands. **When to choose:** Choose this when you need multiple AI assistants to collaborate across different platforms without complex setup. **When NOT to choose:** Avoid if you need high-security communication as the authentication is optional and basic. ### [agent-twitter-client-mcp](https://mcpcatalogs.com/en/server/ryanmac-agent-twitter-client-mcp) by ryanmac · score 44 · ★28 A comprehensive MCP server for interacting with Twitter, supporting various operations including tweet management, user interactions, and Grok AI integration. **When to choose:** Choose this server when you need AI agents to interact with Twitter/Grok for social media monitoring, content creation, or conversation analysis. **When NOT to choose:** Don't choose this if you need direct Twitter API access or require support for non-X Twitter-like platforms. ### [outlook-mcp](https://mcpcatalogs.com/en/server/xenoxilus-outlook-mcp) by XenoXilus · score 43 · ★23 MCP server for Microsoft Outlook enabling AI assistants to access email, calendar, and SharePoint via Microsoft Graph API. **When to choose:** Choose this if you're already using Microsoft 365/O365 and want to integrate Outlook functionality into AI assistants without additional authentication layers. **When NOT to choose:** Avoid if you need access to non-Microsoft email systems or if you require more granular control over authentication flows beyond Azure AD. ### [phone-mcp](https://mcpcatalogs.com/en/server/hao-cyber-phone-mcp) by hao-cyber · score 43 · ★231 Android phone control MCP server using ADB commands for calls, messages, apps, and UI automation. **When to choose:** Choose Phone MCP when you need automated interaction with Android devices for communication, contact management, or UI testing, especially when you're already familiar with ADB or working in a Python environment. **When NOT to choose:** Avoid Phone MCP if you need iOS support, don't want to deal with ADB setup, or require more advanced device management features like app installation or system-level modifications. ### [mcp-twikit](https://mcpcatalogs.com/en/server/adhikasp-mcp-twikit) by adhikasp · score 43 · ★233 MCP server for Twitter integration with search, timeline, and sentiment analysis capabilities. **When to choose:** Choose this when you need to analyze Twitter sentiment or monitor mentions without leaving your AI environment, especially if you're already using Claude Desktop. **When NOT to choose:** Avoid this if you need to post tweets or interact with Twitter beyond reading, as it appears to be read-only only. ### [apple-mail-mcp](https://mcpcatalogs.com/en/server/imdinu-apple-mail-mcp) by imdinu · score 43 · ★24 Apple Mail MCP server with reliable full-text search on large mailboxes, 8 tools for email management. **When to choose:** Choose this for reliable full-text search in large Apple Mail mailboxes where other MCP servers timeout or fail to provide complete coverage. **When NOT to choose:** Not suitable for Windows or Linux users as it's macOS-only, and avoid if you need write operations to your email account. ### [claude-mesh](https://mcpcatalogs.com/en/server/pouriamrt-claude-mesh) by pouriamrt · score 43 · ★22 Networked Claude-to-Claude messaging system enabling direct messaging, broadcasts, threaded conversations and permission relays via MCP channels and self-hosted relay. **When to choose:** Choose claude-mesh when your team uses Claude Code and needs to communicate between instances on different machines without switching to chat applications. **When NOT to choose:** Don't choose this if you need message persistence beyond active sessions or if your team doesn't already use Claude Code for collaboration. ### [mcp-mail](https://mcpcatalogs.com/en/server/shuakami-mcp-mail) by shuakami · score 42 · ★48 MCP-based email management tool enabling AI to handle various email operations including sending, receiving, and organizing emails. **When to choose:** Choose this MCP server when you need AI-powered email management capabilities with support for both reading and composing emails, especially if you're already using an AI assistant that supports the MCP protocol. **When NOT to choose:** Don't choose this if you need advanced calendar integration or task management features beyond email, or if you're using a non-MCP compatible AI assistant without the ability to add custom servers. ### [matrix-mcp-server](https://mcpcatalogs.com/en/server/mjknowles-matrix-mcp-server) by mjknowles · score 42 · ★44 Matrix MCP Server provides secure OAuth 2.0 authentication and 15 tools for Matrix home server integration with room management, messaging, and user profile operations. **When to choose:** Choose this server if you need AI integration with Matrix communication protocols and want to automate room management, messaging, and user interactions within a Matrix ecosystem. **When NOT to choose:** Don't choose this if you need support for non-Matrix chat platforms or if you're looking for a solution that doesn't require OAuth configuration for production deployments. ### [google-mcp](https://mcpcatalogs.com/en/server/vakharwalad23-google-mcp) by vakharwalad23 · score 42 · ★16 MCP server providing access to Google services like Gmail, Calendar, Drive, and Tasks with OAuth authentication. **When to choose:** Choose this server if you need AI integration with Google Workspace services and want seamless OAuth token management with automatic refresh capabilities. **When NOT to choose:** Avoid this if you need access to other Google services not yet implemented (like Google Contacts) or if you prefer a multi-cloud solution beyond Google Workspace. ### [guildbridge](https://mcpcatalogs.com/en/server/dend-guildbridge) by dend · score 42 · ★15 A remote MCP server for Discord that allows AI agents to read, search, and post messages in Discord servers with authenticated access. **When to choose:** Choose GuildBridge when you need AI agents to participate in existing Discord communities and want proper authentication and permission controls. **When NOT to choose:** Don't choose this if you need extensive write permissions beyond basic messaging or if you prefer a self-hosted solution without Cloudflare dependency. ### [better-email-mcp](https://mcpcatalogs.com/en/server/n24q02m-better-email-mcp) by n24q02m · score 42 · ★16 IMAP/SMTP email server for AI agents with 6 composite tools, multi-account support, and auto-discovery. **When to choose:** Choose this server when you need a comprehensive email management solution for AI agents with multi-account support and various email providers, especially if you need both stdio and HTTP deployment options. **When NOT to choose:** Avoid if you need only basic email functionality, as this might be overkill, or if you require end-to-end encryption support which is not explicitly mentioned. ### [ntfy-mcp-server](https://mcpcatalogs.com/en/server/cyanheads-ntfy-mcp-server) by cyanheads · score 42 · ★16 A robust MCP server for sending, managing and replaying ntfy push notifications via STDIO or HTTP transport. **When to choose:** Choose this MCP server if you need to integrate ntfy push notifications into AI workflows, especially when you need granular control over message lifecycle, actions, and filtering options. **When NOT to choose:** Avoid this if you need push notification services beyond ntfy, or if you're looking for a simpler notification solution with fewer configuration options. ### [google-mcp-remote](https://mcpcatalogs.com/en/server/vakharwalad23-google-mcp-remote) by vakharwalad23 · score 41 · ★14 Cloudflare Workers-based MCP server providing Google API tools (Gmail, Calendar, Drive, etc.) for AI clients. **When to choose:** Choose this server if you need AI access to Google services like Gmail, Calendar, Drive, Tasks, YouTube and Contacts, especially when you want a serverless deployment solution without managing your own infrastructure. **When NOT to choose:** Avoid this if you need access to Google services not covered by this server, if you prefer self-hosted solutions over Cloudflare Workers, or if you have concerns about vendor lock-in to Cloudflare's ecosystem. ### [mcp-linkedin](https://mcpcatalogs.com/en/server/adhikasp-mcp-linkedin) by adhikasp · score 41 · ★201 MCP server for LinkedIn API interactions with job search and feed retrieval tools. **When to choose:** Choose this if you need to integrate LinkedIn job searching and feed monitoring into AI workflows without manual browsing. **When NOT to choose:** Avoid if you need official API access, write permissions, or have concerns about using unofficial methods that may break without notice. ### [infobus-mcp](https://mcpcatalogs.com/en/server/simovilab-infobus-mcp) by simovilab · score 41 · ★93 MCP server for transit information access through standardized interfaces. **When to choose:** Choose this MCP server if you're building AI assistants that need to provide transit information and route planning capabilities. **When NOT to choose:** Avoid this server if you need specialized transit features not covered by the standardized interface or if you're looking for broader transportation data APIs beyond MCP. ### [smartlead-mcp-server](https://mcpcatalogs.com/en/server/leadmagic-smartlead-mcp-server) by LeadMagic · score 40 · ★18 SmartLead MCP server provides complete API access to 116+ cold email automation tools with beautiful React Ink installer. **When to choose:** Choose this if you're already using SmartLead for cold email marketing and want AI assistants to manage campaigns, leads, and analytics through MCP tools. **When NOT to choose:** Don't choose this if you need access to email marketing platforms other than SmartLead, or if you prefer open-source solutions without vendor dependencies. ### [linkedin-mcpserver](https://mcpcatalogs.com/en/server/felipfr-linkedin-mcpserver) by felipfr · score 40 · ★62 A TypeScript MCP server integrating LinkedIn API for profile search, job discovery, and messaging capabilities. **When to choose:** Choose this server if you need AI assistant integration with LinkedIn for recruitment, networking, or job search purposes. **When NOT to choose:** Avoid if you need write access beyond messaging or require integration with other professional networks beyond LinkedIn. ### [mcp-server-email](https://mcpcatalogs.com/en/server/shy2593666979-mcp-server-email) by Shy2593666979 · score 39 · ★76 MCP server for sending emails with attachments and searching files in directories. **When to choose:** Choose this server if you need to integrate email functionality into AI workflows and want to search for attachment files without leaving your LLM environment. **When NOT to choose:** Don't choose this if you need advanced email features like templates, scheduling, or integration with existing email systems beyond basic SMTP. ### [protonmail-mcp](https://mcpcatalogs.com/en/server/amotivv-protonmail-mcp) by amotivv · score 39 · ★41 Protonmail MCP server enables email sending through Protonmail's SMTP service for Claude and Cline. **When to choose:** Choose this MCP server if you're already using Protonmail and need to integrate email sending capabilities into Claude Desktop or Cline workflows. **When NOT to choose:** Don't choose this if you need to send emails from a different email provider or require more advanced email features like attachments or templating. ### [Faxbot](https://mcpcatalogs.com/en/server/dmontgomery40-faxbot) by DMontgomery40 · score 38 · ★7 A self-hosted fax API with MCP support for AI assistant integration. **When to choose:** Choose Faxbot if you need self-hosted fax capabilities with AI assistant integration and already have existing infrastructure for deployment. **When NOT to choose:** Don't choose Faxbot if you need a simple cloud fax service without self-hosting requirements, or if you require extensive compliance certifications beyond the HIPAA-aligned controls it provides. ### [tgfmcp](https://mcpcatalogs.com/en/server/vaibhavpandeyvpz-tgfmcp) by vaibhavpandeyvpz · score 38 · ★2 Telegram stdio MCP server exposing Bot API tools and message notifications through Telegraf. **When to choose:** Choose tgfmcp when you need to integrate Telegram messaging into AI workflows through MCP, especially for building chatbots or notification systems with full message control capabilities. **When NOT to choose:** Don't use tgfmcp if you require bidirectional communication with Telegram users beyond basic bot functionality, as bots cannot initiate conversations or access arbitrary private chats. ### [mastodon-mcp](https://mcpcatalogs.com/en/server/the-focus-ai-mastodon-mcp) by The-Focus-AI · score 38 · ★9 MCP server for Mastodon social media interaction with toot creation and media attachments. **When to choose:** Choose this MCP server if you need to programmatically interact with Mastodon, especially for scheduling posts or managing media attachments with accessibility descriptions. **When NOT to choose:** Don't choose this server if you need to manage multiple accounts or interact with other social media platforms beyond Mastodon, as it's specialized for a single platform. ### [mcp-notifications](https://mcpcatalogs.com/en/server/zudsniper-mcp-notifications) by zudsniper · score 37 · ★6 MCP server for real-time notifications and webhooks when AI agents complete tasks. **When to choose:** Choose this when you're working with multiple AI agent projects and need efficient task completion notifications without switching contexts. **When NOT to choose:** Avoid if you need a self-hosted solution without external dependencies (Cloudflare Workers, PocketBase) or if you're already heavily invested in another notification system. ### [titanmind-whatsapp-mcp](https://mcpcatalogs.com/en/server/titanmindagi-titanmind-whatsapp-mcp) by TitanmindAGI · score 37 · ★18 WhatsApp marketing and messaging MCP server using Titanmind for conversation management and template-based messaging. **When to choose:** Choose this MCP if you need WhatsApp marketing capabilities and are already using or willing to integrate with Titanmind's services. **When NOT to choose:** Avoid if you need direct WhatsApp API access without dependency on Titanmind's platform or if you need messaging outside of WhatsApp. ### [wappmcp](https://mcpcatalogs.com/en/server/vaibhavpandeyvpz-wappmcp) by vaibhavpandeyvpz · score 36 · ★3 WhatsApp Web MCP server enabling message management and event notifications through stdio. **When to choose:** Choose this when you need to integrate WhatsApp functionality into AI workflows via MCP and want to manage messages programmatically. **When NOT to choose:** Avoid if you need official WhatsApp API access, enterprise-grade reliability, or if the initial QR code authentication is impractical for your use case. ### [phonepi-mcp](https://mcpcatalogs.com/en/server/priyankark-phonepi-mcp) by priyankark · score 35 · ★34 PhonePi MCP provides 23+ tools to control your smartphone remotely from AI assistants. **When to choose:** Choose PhonePi MCP when you need AI integration with Android-specific phone features and are comfortable with local network deployment. **When NOT to choose:** Don't choose this if you need iOS support, cloud-based remote access without local network setup, or enterprise-grade security features. ### [gmail-mcp](https://mcpcatalogs.com/en/server/hahaha-saygex-gmail-mcp) by hahaha-saygex · score 34 · ★1 Gmail MCP server for email management with OAuth2 authentication. **When to choose:** Choose this if you need a prebuilt solution for Gmail management and prefer binaries over source code implementation. **When NOT to choose:** Avoid this if you need source code visibility, want to understand or customize the MCP implementation, or need clear documentation on MCP integration. ### [mcp-walichat](https://mcpcatalogs.com/en/server/walichat-mcp-walichat) by walichat · score 34 · ★1 WhatsApp MCP connector for AI clients to automate messaging, analyze conversations, and manage groups via WaliChat API. **When to choose:** Choose this MCP server if you need to integrate WhatsApp automation with AI assistants and want comprehensive messaging, group management, and analytics capabilities through WaliChat. **When NOT to choose:** Don't choose this if you need to work with a different messaging platform, or if you're not willing to use a third-party API that requires a paid subscription for full functionality. ### [mcp-email-client](https://mcpcatalogs.com/en/server/gamalan-mcp-email-client) by gamalan · score 34 · ★17 Python-based email client MCP server for managing multiple email configurations and operations. **When to choose:** Choose this MCP server when you need to integrate email operations into Claude Desktop workflows, especially if you use multiple email accounts and want basic email management capabilities. **When NOT to choose:** Don't choose this server if you need advanced email features, detailed documentation about supported providers, or require more than the latest 5 unread emails. ### [Java-MCP-Server-For-SMTP-Mailing](https://mcpcatalogs.com/en/server/rayenmalouche-java-mcp-server-for-smtp-mailing) by RayenMalouche · score 32 · ★2 Java-based MCP server providing SMTP email sending capabilities via multiple transport protocols. **When to choose:** Choose this server if you're building Java-based AI applications that need email capabilities and require integration with Claude Desktop or other MCP clients. **When NOT to choose:** Don't choose this if you need HTML email formatting, attachment support, or a non-Java implementation; these features are either missing or planned but not yet implemented. ### [elasticemail-mcp-server](https://mcpcatalogs.com/en/server/elasticemail-elasticemail-mcp-server) by ElasticEmail · score 31 · ★1 ElasticEmail MCP server enables AI agents to send, manage, and track email campaigns through a full suite of email operations. **When to choose:** Choose this if you're already using Elastic Email and want AI agents to manage your email campaigns, contacts, and analytics without switching platforms. **When NOT to choose:** Don't choose this if you need email services outside of Elastic Email, as this creates vendor lock-in to their platform and API. ### [buttondown-mcp](https://mcpcatalogs.com/en/server/the-focus-ai-buttondown-mcp) by The-Focus-AI · score 31 · ★0 MCP server for Buttondown newsletter service with CLI and programmatic API. **When to choose:** Choose this MCP server if you're already using Buttondown for newsletters and want AI/LLM integration to streamline content creation, scheduling, and analytics. **When NOT to choose:** Don't choose this if you need to integrate with a different newsletter service or require extensive customization beyond Buttondown's API capabilities. ### [twitter-mcp](https://mcpcatalogs.com/en/server/dishant27-twitter-mcp) by Dishant27 · score 30 · ★3 MCP server for X (Twitter) integration enabling CRUD operations via Twitter API. **When to choose:** Choose this server for Twitter/X integration when you need a straightforward MCP interface for basic operations like posting, searching, and account management. **When NOT to choose:** Don't choose this if you need advanced Twitter features like analytics, API v2 endpoints not covered here, or if you require enterprise-level access controls. ### [telegram-mcp-server](https://mcpcatalogs.com/en/server/abhinavkale-dev-telegram-mcp-server) by abhinavkale-dev · score 30 · ★4 A Model Context Protocol server enabling AI assistants to interact with Telegram through standardized tools. **When to choose:** Choose this when you need basic Telegram integration for AI assistants and want a simple, protocol-based solution without complex setup. **When NOT to choose:** Don't choose this if you need advanced Telegram features like group management, file uploads beyond photos, or require more robust error handling. --- ## database (74) ### [dbhub](https://mcpcatalogs.com/en/server/bytebase-dbhub) by bytebase · score 57 · ★2,769 A zero-dependency MCP server connecting to multiple databases with token-efficient SQL tools. **When to choose:** Choose DBHub when you need token-efficient database access across multiple systems with built-in safety controls and prefer a zero-dependency installation. **When NOT to choose:** Don't choose DBHub if you need write access to databases or require complex transactional operations beyond its read-only capabilities. ### [mcp-server-qdrant](https://mcpcatalogs.com/en/server/qdrant-mcp-server-qdrant) by qdrant · score 55 · ★1,397 Official Qdrant MCP server for semantic memory storage and retrieval using vector embeddings. **When to choose:** Choose this MCP server when you need vector-based semantic memory storage and retrieval for AI applications, especially if you're already using or planning to use Qdrant as your vector database. **When NOT to choose:** Don't choose this if you need a different vector database backend, as this server is tightly coupled with Qdrant and doesn't support other storage backends. ### [mongodb-mcp-server](https://mcpcatalogs.com/en/server/mongodb-js-mongodb-mcp-server) by mongodb-js · score 55 · ★1,023 MongoDB MCP Server connects to MongoDB databases and Atlas clusters, offering tools for database operations and Atlas management. **When to choose:** Choose this when working with MongoDB databases or Atlas clusters and need secure AI access to your data without building custom integrations. **When NOT to choose:** Avoid if you need write access by default (must explicitly remove readOnly flag) or if you require support for non-MongoDB databases. ### [pg-aiguide](https://mcpcatalogs.com/en/server/timescale-pg-aiguide) by timescale · score 55 · ★1,729 MCP server providing PostgreSQL expertise, semantic search, and best practices for AI coding tools. **When to choose:** Choose pg-aiguide when working with PostgreSQL and need AI coding assistance to follow modern best practices, generate proper database schemas with constraints and indexes, and search PostgreSQL documentation within your development environment. **When NOT to choose:** Don't choose pg-aiguide if you primarily work with non-PostgreSQL databases, as it's specifically tailored for PostgreSQL expertise only. ### [restheart](https://mcpcatalogs.com/en/server/softinstigate-restheart) by SoftInstigate · score 54 · ★875 RESTHeart is a MongoDB backend that provides a native MCP server for AI agents to access data via REST and GraphQL APIs. **When to choose:** Choose RESTHeart when you need a production-ready backend for AI agents to access MongoDB securely without custom integration code. **When NOT to choose:** Avoid if you need AGPLv3 compliance for all components (plugins have Apache 2.0 exception but core is AGPLv3) or if you require write capabilities (this is primarily a read-only MCP server). ### [mcp-redis](https://mcpcatalogs.com/en/server/redis-mcp-redis) by redis · score 54 · ★510 Official Redis MCP Server providing natural language interface for AI agents to manage Redis data efficiently. **When to choose:** Choose this when you need an official, production-ready MCP interface for Redis with comprehensive data structure support and natural language capabilities. **When NOT to choose:** Avoid if you're not already using Redis or if you need specialized database features not supported by Redis like complex joins or multi-document transactions. ### [dbt-mcp](https://mcpcatalogs.com/en/server/dbt-labs-dbt-mcp) by dbt-labs · score 53 · ★562 A comprehensive MCP server for dbt Core, Fusion, and Platform, providing SQL, semantic layer, and dbt CLI tools. **When to choose:** Choose dbt MCP if you're working with dbt projects and want AI agents to understand your data model structure, execute SQL against your warehouse, or manage dbt operations through natural language. **When NOT to choose:** Don't choose this if you're not using dbt or need write access beyond what dbt CLI provides - it's designed for read analysis and dbt-specific operations rather than general database management. ### [airtable-mcp-server](https://mcpcatalogs.com/en/server/domdomegg-airtable-mcp-server) by domdomegg · score 53 · ★444 A comprehensive MCP server enabling AI systems to interact with Airtable databases through read and write operations. **When to choose:** Choose this when you need full read/write access to Airtable bases through AI assistants and want a well-maintained, actively developed server with comprehensive tool coverage. **When NOT to choose:** Don't choose this if you need access to Airtable's automation interfaces (like interfaces) or if you require authentication beyond the API token for secure environments. ### [data-api-builder](https://mcpcatalogs.com/en/server/azure-data-api-builder) by Azure · score 53 · ★1,410 Azure's Data API Builder provides REST and GraphQL endpoints for Azure databases with upcoming MCP tools support. **When to choose:** Choose Data API Builder when you need to expose multiple database types through APIs and want Microsoft's tooling integration, especially if you're already using Azure services. **When NOT to choose:** Avoid if you need immediate MCP integration (it's coming soon but not available yet) or if you require write access beyond CRUD operations, as DAB focuses primarily on data access patterns. ### [redd-archiver](https://mcpcatalogs.com/en/server/19-84-redd-archiver) by 19-84 · score 52 · ★333 A PostgreSQL-backed archive generator that creates browsable HTML archives from link aggregator platforms with MCP server integration. **When to choose:** Choose this when you need AI access to historical forum data from Reddit, Voat, or Ruqqus, especially if you're already using PostgreSQL. **When NOT to choose:** Don't choose this if you need write access to the database (it's read-only) or if you require real-time data from active platforms. ### [mcp-neo4j](https://mcpcatalogs.com/en/server/neo4j-contrib-mcp-neo4j) by neo4j-contrib · score 52 · ★947 Neo4j Labs MCP servers integrate graph databases with AI assistants through natural language. **When to choose:** Choose these MCP servers if you work with Neo4j graph databases and want to integrate them directly into AI workflows for querying, knowledge management, or cloud instance management. **When NOT to choose:** Avoid if you don't use Neo4j, as these servers provide limited functionality without a Neo4j instance, and note that they are experimental without SLA or backward compatibility guarantees. ### [mysql_mcp_server](https://mcpcatalogs.com/en/server/designcomputer-mysql-mcp-server) by designcomputer · score 52 · ★1,251 MCP server enabling secure interaction with MySQL databases through structured API. **When to choose:** Choose this when you need secure, structured access to MySQL databases for AI applications with comprehensive logging and environment variable-based security. **When NOT to choose:** Don't choose this if you need write access to databases (current implementation appears read-only) or if you require authentication beyond environment variables. ### [db-mcp-server](https://mcpcatalogs.com/en/server/freepeak-db-mcp-server) by FreePeak · score 52 · ★377 Multi-database MCP server connecting AI assistants to MySQL, PostgreSQL, SQLite and Oracle through standardized SQL interfaces. **When to choose:** Teams working with multiple database systems (MySQL, PostgreSQL, SQLite, Oracle) who need AI assistants to analyze data across different databases without direct database access. **When NOT to choose:** If you need write access to databases (this server appears focused on read operations), or if you're working exclusively with databases not in the supported list. ### [NornicDB](https://mcpcatalogs.com/en/server/orneryd-nornicdb) by orneryd · score 52 · ★734 NornicDB is a distributed graph+vector database with native MCP server support for AI applications. **When to choose:** Choose NornicDB when you need hybrid graph+vector queries with temporal consistency in a single database, particularly for Graph-RAG systems, agent memory, or knowledge graphs requiring audit trails. **When NOT to choose:** Avoid NornicDB if you need extensive graph visualization tools, have existing Neo4j bolt extensions that aren't compatible, or prefer a managed cloud service over self-hosted solutions. ### [MCP-Salesforce](https://mcpcatalogs.com/en/server/smn2gnt-mcp-salesforce) by smn2gnt · score 50 · ★178 MCP Salesforce Connector enables LLMs to interact with Salesforce data through SOQL queries and SOSL searches. **When to choose:** Choose this MCP server when you need to integrate AI/LLMs with Salesforce CRM data and want comprehensive CRUD operations plus advanced API access. **When NOT to choose:** Avoid if you don't already use Salesforce, as it requires Salesforce authentication credentials and deep knowledge of Salesforce APIs. ### [mcp-mongo-server](https://mcpcatalogs.com/en/server/kiliczsh-mcp-mongo-server) by kiliczsh · score 50 · ★278 An MCP server for MongoDB interaction with schema inference, query capabilities, and configurable read-only mode. **When to choose:** Choose this when you need an MCP interface for MongoDB and want fine-grained control over read/write access, especially with its unique ObjectId handling capabilities. **When NOT to choose:** Avoid if you need authentication beyond basic URI connection or require support for non-MongoDB databases. ### [MCP-PostgreSQL-Ops](https://mcpcatalogs.com/en/server/call518-mcp-postgresql-ops) by call518 · score 50 · ★149 Comprehensive MCP server providing PostgreSQL DBA superpowers with 30+ tools for performance analysis, bloat detection, and monitoring. **When to choose:** Choose this for PostgreSQL DBA operations when you need comprehensive monitoring, performance analysis, and maintenance recommendations through natural language queries, especially with production constraints requiring read-only access. **When NOT to choose:** Don't choose this if you need write access to PostgreSQL databases, as this server is read-only only. Also consider alternatives if you require monitoring for non-PostgreSQL databases. ### [PerformanceStudio](https://mcpcatalogs.com/en/server/erikdarlingdata-performancestudio) by erikdarlingdata · score 50 · ★184 A cross-platform SQL Server execution plan analyzer with built-in MCP server for AI-assisted analysis. **When to choose:** Choose Performance Studio when you need AI-assisted SQL Server execution plan analysis through MCP integration, especially if you're already working with SQL Server performance optimization. **When NOT to choose:** Not for PostgreSQL or Oracle databases; limited to SQL Server. Also not suitable if you need to analyze execution plans from non-Microsoft database systems. ### [tiger-cli](https://mcpcatalogs.com/en/server/timescale-tiger-cli) by timescale · score 49 · ★108 Command-line interface for Tiger Cloud with MCP server for AI assistants to manage Postgres services and execute queries. **When to choose:** Choose this when working with TimescaleDB on Tiger Cloud and you want AI assistants to manage services and query databases without leaving your coding environment. **When NOT to choose:** Don't choose this if you need write access to databases (it's read-only by default) or if you're looking for a vendor-agnostic solution for PostgreSQL management. ### [pgmcp](https://mcpcatalogs.com/en/server/subnetmarco-pgmcp) by subnetmarco · score 49 · ★529 PGMCP is a production-ready MCP server that enables natural language queries to any PostgreSQL database with proper security controls. **When to choose:** Choose PGMCP when you need to provide AI assistants with read-only access to existing PostgreSQL databases without modifying your schema or writing SQL. **When NOT to choose:** Don't choose PGMCP if you need write access to your database or require AI-generated modifications to your schema. ### [mcp-server-elasticsearch](https://mcpcatalogs.com/en/server/elastic-mcp-server-elasticsearch) by elastic · score 49 · ★662 Deprecated Elasticsearch MCP server for AI agents to query and analyze data. **When to choose:** Only for legacy systems using older Elastic versions before 9.2.0 that cannot upgrade to Elastic Agent Builder. **When NOT to choose:** Should not be chosen as it's deprecated - Elastic recommends using Agent Builder for new implementations. ### [airtable-mcp](https://mcpcatalogs.com/en/server/rashidazarang-airtable-mcp) by rashidazarang · score 48 · ★80 A comprehensive MCP server for Airtable integration with full CRUD operations, schema management, and AI-powered analytics. **When to choose:** Choose this if you need comprehensive Airtable integration with AI-powered analytics and governance features, especially if already using Airtable in your workflow. **When NOT to choose:** Avoid if you need read/write access to other data sources beyond Airtable, or if you're looking for lightweight integration without advanced governance features. ### [surrealmcp](https://mcpcatalogs.com/en/server/surrealdb-surrealmcp) by surrealdb · score 48 · ★94 Official SurrealDB MCP server enabling AI assistants to interact with SurrealDB databases. **When to choose:** Choose SurrealMCP when you need AI tools to interact with SurrealDB databases and want official support with multiple transport options. **When NOT to choose:** Avoid SurrealMCP if you're already deeply invested in another database ecosystem, as this creates vendor lock-in to SurrealDB. ### [mcp-bigquery-server](https://mcpcatalogs.com/en/server/ergut-mcp-bigquery-server) by ergut · score 48 · ★139 A secure MCP server enabling LLMs to query BigQuery datasets with field-level data protection. **When to choose:** Choose this MCP server if you're already using Google BigQuery and need AI-powered data access with sensitive field protection capabilities. **When NOT to choose:** Don't choose this if you need write access to BigQuery or are looking for a database-agnostic solution that works with multiple cloud data warehouses. ### [powerbi-mcp](https://mcpcatalogs.com/en/server/sulaiman013-powerbi-mcp) by sulaiman013 · score 48 · ★111 An enterprise-grade MCP server for Power BI that enables AI assistants to interact with datasets via natural language queries, supporting both desktop and cloud environments. **When to choose:** Choose this server when you need AI assistants to interact with Power BI datasets through natural language, especially when working with complex model refactoring while preserving report visual integrity. **When NOT to choose:** Avoid this if you're not already using Power BI, need write access to datasets (this primarily focuses on read/analysis operations), or require cross-platform support (Windows-only). ### [mcp-metabase-server](https://mcpcatalogs.com/en/server/easecloudio-mcp-metabase-server) by easecloudio · score 48 · ★76 A comprehensive MCP server for Metabase providing 96 tools to manage dashboards, cards, databases, tables, and more. **When to choose:** Choose this server when you need comprehensive AI access to your Metabase instance with full CRUD operations across dashboards, cards, databases, and tables. **When NOT to choose:** Avoid if you need access to non-Metabase data sources or require read-only access without the ability to create/update/delete content through AI. ### [samyama-graph](https://mcpcatalogs.com/en/server/samyama-ai-samyama-graph) by samyama-ai · score 48 · ★62 High-performance graph-vector database with MCP server support for natural language queries over knowledge graphs. **When to choose:** Choose Samyama when working with massive graph datasets (1B+ edges) that need both traditional graph operations and vector search, particularly in biomedical research or enterprise knowledge management. **When NOT to choose:** Avoid Samyama if you need a mature cloud-hosted solution with enterprise-grade security features, as it's a self-hosted system with limited multi-tenant access controls and auditing. ### [mcp-server-starrocks](https://mcpcatalogs.com/en/server/starrocks-mcp-server-starrocks) by StarRocks · score 48 · ★171 Official MCP server connecting AI assistants to StarRocks databases for SQL execution, data exploration, and visualization. **When to choose:** Choose this server when working with StarRocks databases and need AI assistants to perform SQL queries, explore schemas, and create data visualizations without complex client setup. **When NOT to choose:** Don't choose this server if you need to connect to non-StarRocks databases or require advanced features not implemented by this server. ### [mcp-trino](https://mcpcatalogs.com/en/server/tuannvm-mcp-trino) by tuannvm · score 48 · ★103 A high-performance Go implementation of MCP server for Trino SQL query engine with dual CLI and MCP modes. **When to choose:** Choose this MCP server when you need Go performance for Trino integration, require both MCP and CLI modes, or need robust OAuth authentication for enterprise environments. **When NOT to choose:** Don't choose this if you're already invested in a different language ecosystem or need to connect to non-Trino data sources directly. ### [ChEMBL-MCP-Server](https://mcpcatalogs.com/en/server/augmented-nature-chembl-mcp-server) by Augmented-Nature · score 48 · ★83 A comprehensive MCP server providing 22 specialized tools for ChEMBL chemical database access in drug discovery research. **When to choose:** Choose this MCP server when you need AI agents to perform advanced chemical research, drug discovery, or bioinformatics analysis using the comprehensive ChEMBL database. **When NOT to choose:** Avoid using this server if you need write access to chemical data or require offline functionality, as it only provides read-only access via ChEMBL's REST API. ### [metabase-mcp](https://mcpcatalogs.com/en/server/hluaguo-metabase-mcp) by hluaguo · score 47 · ★64 Metabase MCP server enables AI assistants to query databases, execute SQL, and manage dashboards through Metabase API. **When to choose:** Choose this server when you're already using Metabase for analytics and want AI assistants to directly query your databases through their API without needing additional data preparation. **When NOT to choose:** Don't choose this if you need direct database access outside of Metabase or if your Metabase instance doesn't have API access enabled. ### [mcp-database-server](https://mcpcatalogs.com/en/server/executeautomation-mcp-database-server) by executeautomation · score 47 · ★351 MCP Database Server connects SQLite, SQL Server, PostgreSQL and MySQL databases to Claude. **When to choose:** Choose this MCP server when you need unified access to multiple database types (SQLite, SQL Server, PostgreSQL, MySQL) through Claude without writing custom integrations. **When NOT to choose:** Avoid if you need specialized database features not covered by basic SQL operations or require enterprise-level security controls beyond what's documented. ### [hana-mcp-server](https://mcpcatalogs.com/en/server/hatrigt-hana-mcp-server) by HatriGt · score 47 · ★54 SAP HANA MCP server enables AI clients to query HANA databases with guardrails and merge business metadata. **When to choose:** Choose this server if you need to integrate AI assistants with SAP HANA databases, especially when you require SQL guardrails and business metadata enrichment. **When NOT to choose:** Avoid this if you're not using SAP HANA databases or need write capabilities beyond SQL query execution. ### [mcp-server-bigquery](https://mcpcatalogs.com/en/server/lucashild-mcp-server-bigquery) by LucasHild · score 47 · ★126 An MCP server enabling LLMs to inspect BigQuery schemas and execute SQL queries. **When to choose:** Teams using Google Cloud Platform's BigQuery who want AI assistants to analyze their data without requiring direct console access. **When NOT to choose:** You need write access to BigQuery (this appears read-only), or if you're not already using Google Cloud Platform. ### [mssql-mcp](https://mcpcatalogs.com/en/server/aaronontheweb-mssql-mcp) by Aaronontheweb · score 47 · ★145 A .NET-powered MCP server for Microsoft SQL Server with robust tools for schema discovery, query execution, and connection validation. **When to choose:** Choose this when working specifically with Microsoft SQL Server and need a reliable, production-ready MCP server with robust .NET implementation and Docker support. **When NOT to choose:** Not ideal if you need cross-database support beyond SQL Server or prefer a language-agnostic solution not tied to .NET ecosystem. ### [firebase-mcp](https://mcpcatalogs.com/en/server/gannonh-firebase-mcp) by gannonh · score 46 · ★243 Firebase MCP server providing Firestore, Storage, and Authentication tools for AI assistants. **When to choose:** Choose Firebase MCP when you need AI assistants to directly interact with Firebase services, especially if you're already using Firebase in your application stack and want to eliminate the need for custom API integrations. **When NOT to choose:** Avoid this server if you don't use Firebase services or if you need more extensive database operations beyond what Firebase offers, as it's locked into Firebase's ecosystem. ### [us-census-bureau-data-api-mcp](https://mcpcatalogs.com/en/server/uscensusbureau-us-census-bureau-data-api-mcp) by uscensusbureau · score 46 · ★71 MCP server connecting AI assistants with official U.S. Census Bureau statistics and data. **When to choose:** Choose this MCP server when you need authoritative U.S. demographic and economic data in your AI applications, especially if you're already using Docker and want direct access to Census Bureau statistics without building custom integrations. **When NOT to choose:** Don't choose this if you need data for countries other than the U.S., or if you prefer a lightweight solution without Docker dependency. The server also requires a Census API key which requires registration with the Census Bureau. ### [mysql_mcp_server_pro](https://mcpcatalogs.com/en/server/wenb1n-dev-mysql-mcp-server-pro) by wenb1n-dev · score 46 · ★244 A comprehensive MCP server for MySQL databases with advanced query, analysis, and optimization capabilities. **When to choose:** Choose this MySQL MCP server when you need database health monitoring, performance optimization, and Chinese field name support in addition to basic CRUD operations. **When NOT to choose:** Don't choose this if you need support for non-MySQL databases, or if you require a simpler solution with fewer specialized features. ### [flapi](https://mcpcatalogs.com/en/server/datazoode-flapi) by DataZooDE · score 46 · ★70 flAPI generates SQL-based APIs with DuckDB and provides MCP tools for AI agents. **When to choose:** Choose flAPI when you need to quickly create APIs for existing data warehouses or databases that AI agents need to access, especially when you're already using DuckDB or want read-only access without ETL. **When NOT to choose:** Don't choose flAPI if you need write access to your databases, require complex business logic beyond SQL queries, or need to connect to data sources not supported by DuckDB. ### [mcp-server](https://mcpcatalogs.com/en/server/keboola-mcp-server) by keboola · score 46 · ★84 Keboola MCP Server bridges AI agents with Keboola platform, exposing data, SQL queries, and job triggers as MCP tools. **When to choose:** Choose this server if you're already using Keboola and want to integrate it directly with AI agents without writing custom integration code. **When NOT to choose:** Avoid this if you need write access beyond job triggering or if you're not already invested in the Keboola ecosystem. ### [nocodb-mcp-server](https://mcpcatalogs.com/en/server/edwinbernadus-nocodb-mcp-server) by edwinbernadus · score 46 · ★70 NocoDB MCP Server enables CRUD operations on NocoDB databases through natural language commands. **When to choose:** Choose this server if you're already using NocoDB and want to interact with it through conversational AI interfaces or MCP-enabled tools. **When NOT to choose:** Avoid this if you need access to database types other than NocoDB, or if you require more advanced database operations beyond basic CRUD. ### [orionbelt-analytics](https://mcpcatalogs.com/en/server/ralfbecher-orionbelt-analytics) by ralfbecher · score 45 · ★29 Ontology-based MCP server analyzing database schemas and generating RDF/OWL ontologies with SQL mappings for Text-to-SQL. **When to choose:** Choose OrionBelt when working with complex relational databases where data relationships aren't intuitive and you need deterministic validation of generated SQL. **When NOT to choose:** Don't choose OrionBelt if you need write access to databases (it's read-only) or if you prefer a simpler SQL generation approach without ontology complexity. ### [bonnard](https://mcpcatalogs.com/en/server/bonnard-data-bonnard) by bonnard-data · score 44 · ★26 Bonnard is an agent-native semantic layer providing MCP server capabilities for reliable data querying through AI agents. **When to choose:** Choose Bonnard when you need a self-hosted semantic layer for AI agents with existing database infrastructure and prefer not to use cloud services. **When NOT to choose:** Don't choose Bonnard if you need real-time data updates (it's optimized for analytical queries) or if you prefer a fully managed cloud service over self-hosting. ### [apisql-mcp](https://mcpcatalogs.com/en/server/apisql-dev-apisql-mcp) by apisql-dev · score 44 · ★33 MCP server providing unified access to multiple databases through SQL operations with dynamic switching capabilities. **When to choose:** When you need to work with multiple database types through AI assistants and want a unified interface without setting up separate MCP servers for each database. **When NOT to choose:** When you need direct database access without depending on the apiSQL platform or when working in an environment without internet connectivity to the apiSQL service. ### [Augmented-Nature-UniProt-MCP-Server](https://mcpcatalogs.com/en/server/augmented-nature-augmented-nature-uniprot-mcp-server) by Augmented-Nature · score 44 · ★19 Comprehensive UniProt MCP server with 26 bioinformatics tools for protein research and analysis. **When to choose:** Choose this server when you need specialized bioinformatics tools for protein research, sequence analysis, or evolutionary studies directly within your MCP-enabled AI environment. **When NOT to choose:** Don't choose this server if you need write access to protein databases or require integration with other biological databases not supported by the UniProt API. ### [go-mcp-mysql](https://mcpcatalogs.com/en/server/zhwt-go-mcp-mysql) by Zhwt · score 44 · ★53 A production-ready Go-based MCP server for MySQL database operations with read-only mode and query plan checking. **When to choose:** Choose this server when you need a lightweight MySQL integration for AI systems without Node.js/Python dependencies, or when you specifically need query plan validation via EXPLAIN statements. **When NOT to choose:** Avoid this if you need advanced security features, authentication beyond basic credentials, or production stability as it's marked as work in progress. ### [pocketbase-mcp](https://mcpcatalogs.com/en/server/mrwyndham-pocketbase-mcp) by mrwyndham · score 44 · ★134 MCP server for PocketBase databases with CRUD operations, schema management, and backup capabilities. **When to choose:** Choose this MCP server when working with PocketBase databases and want AI assistance with database operations, schema management, and CRUD operations through MCP. **When NOT to choose:** Don't choose this if you need support for other database backends or if you require features not yet implemented in this MCP server. ### [turbular](https://mcpcatalogs.com/en/server/raeudigerraeffi-turbular) by raeudigerRaeffi · score 44 · ★99 Turbular is an MCP server that enables LLM agents to connect to and query multiple databases through a unified API interface. **When to choose:** Choose Turbular when you need an MCP solution to connect LLM agents to multiple database types with minimal configuration and automatic schema normalization. **When NOT to choose:** Don't choose Turbular if you need write access to databases (current implementation appears read-only) or require advanced authentication beyond what's documented. ### [mongodb-lens](https://mcpcatalogs.com/en/server/furey-mongodb-lens) by furey · score 43 · ★201 MongoDB Lens is a full-featured MCP server enabling natural language interaction with MongoDB databases. **When to choose:** Choose MongoDB Lens if you need comprehensive natural language access to MongoDB databases without switching between tools and interfaces. **When NOT to choose:** Avoid if you need multi-database support beyond MongoDB or require write access to databases that should remain untouched. ### [SmartDB_MCP](https://mcpcatalogs.com/en/server/wenb1n-dev-smartdb-mcp) by wenb1n-dev · score 43 · ★79 Universal database MCP server connecting to MySQL, PostgreSQL, SQL Server, MariaDB, DM8, Oracle with OAuth2 authentication, health checks, SQL optimization, and index health detection. **When to choose:** Choose SmartDB when you need a unified interface to multiple database systems with advanced features like SQL optimization and health monitoring, especially when working with read-only access scenarios. **When NOT to choose:** Don't choose SmartDB if you need frequent write access or administrative privileges, as it primarily focuses on read-only and read-write scenarios with limited admin capabilities. ### [data-gov-il-mcp](https://mcpcatalogs.com/en/server/davidosherdiagnostica-data-gov-il-mcp) by DavidOsherdiagnostica · score 43 · ★64 Advanced MCP server providing access to Israeli government open datasets with Hebrew/English search capabilities. **When to choose:** Choose this MCP server when you need access to Israeli government open data and want to integrate it directly into AI workflows without additional API management. **When NOT to choose:** Avoid this server if you need write access to government data (it's read-only only) or if you require data from other countries' government sources. ### [mcp-server-duckdb](https://mcpcatalogs.com/en/server/ktanaka101-mcp-server-duckdb) by ktanaka101 · score 42 · ★176 MCP server for DuckDB database interaction, enabling SQL query execution through MCP tools. **When to choose:** Choose this when you need to enable AI models to query local DuckDB databases without requiring specialized database access setup. **When NOT to choose:** Don't choose this if you need write access to the database in multi-user environments, as there's no built-in authentication or concurrent access control. ### [skalex](https://mcpcatalogs.com/en/server/tarekraafat-skalex) by TarekRaafat · score 42 · ★16 Zero-dependency JavaScript database with built-in vector search, agent memory, and MCP server functionality. **When to choose:** Choose Skalex when you need a lightweight, zero-dependency database with built-in AI capabilities and want to expose it through MCP to AI agents. **When NOT to choose:** Don't choose Skalex for large-scale distributed systems or multi-process applications that require traditional database features like ACID transactions across processes. ### [mcp-turso-cloud](https://mcpcatalogs.com/en/server/spences10-mcp-turso-cloud) by spences10 · score 42 · ★15 A well-documented MCP server for Turso database integration with dual authentication and query separation. **When to choose:** Choose this MCP server if you're already using Turso databases and want to enable LLM interactions while maintaining secure separation between read and write operations. **When NOT to choose:** Don't choose this if you need to work with other database systems as it's specifically designed for Turso with limited portability. ### [mcp-dbutils](https://mcpcatalogs.com/en/server/donghao1393-mcp-dbutils) by donghao1393 · score 41 · ★88 mcp-dbutils is a secure MCP server enabling AI to analyze data across multiple database types with read-only access. **When to choose:** Choose this MCP server when you need secure, read-only AI access to multiple database types without exposing credentials directly to AI models. **When NOT to choose:** Avoid this server if you need write access to databases or require direct SQL execution capabilities beyond SELECT queries. ### [legion-mcp](https://mcpcatalogs.com/en/server/theralabs-legion-mcp) by TheRaLabs · score 41 · ★91 A MCP server providing unified database access across PostgreSQL, MySQL, SQL Server and others via the Legion Query Runner. **When to choose:** Choose this MCP server when you need unified access to multiple database types and want to integrate database operations into AI assistants without learning different client libraries. **When NOT to choose:** Avoid this if you need write access to databases or require advanced query optimization features, as this appears to be focused primarily on read operations. ### [lightdash-mcp-server](https://mcpcatalogs.com/en/server/syucream-lightdash-mcp-server) by syucream · score 41 · ★25 MCP server providing AI assistants with standardized access to Lightdash's analytics platform through API tools. **When to choose:** Choose this MCP server if you're already using Lightdash and want to integrate it with AI assistants for natural data exploration and insights generation. **When NOT to choose:** Avoid this if you need data write capabilities, as the server only provides read-only access to Lightdash resources. ### [mysql-mcp-server](https://mcpcatalogs.com/en/server/tickhaijun-mysql-mcp-server) by TickHaiJun · score 41 · ★31 A Node.js MCP server that connects to MySQL databases, providing SQL query execution and metadata retrieval. **When to choose:** Choose this server when you need an MCP interface for MySQL with basic query capabilities and built-in safety features, especially if you're working with JavaScript-based AI systems. **When NOT to choose:** Don't choose this if you need advanced database operations (transactions, stored procedures) or require fine-grained access control beyond basic safety checks. ### [mcp-libsql](https://mcpcatalogs.com/en/server/xexr-mcp-libsql) by Xexr · score 40 · ★20 Secure MCP server for libSQL databases with comprehensive tools, connection pooling, and transaction support. **When to choose:** Choose this server when working with libSQL or Turso databases and need secure, production-ready database access through MCP protocol. **When NOT to choose:** Don't choose this if you need support for database systems other than libSQL, or if you require read/write capabilities beyond what libSQL provides. ### [postgresql-ssh-mcp](https://mcpcatalogs.com/en/server/zlash65-postgresql-ssh-mcp) by Zlash65 · score 39 · ★6 PostgreSQL MCP server with SSH tunneling for secure database access through bastion hosts. **When to choose:** Choose this server when you need secure PostgreSQL access through SSH tunnels, especially for ChatGPT integration or Claude Desktop in restricted environments. **When NOT to choose:** Don't choose this if you need write access by default (it's read-only-only unless explicitly configured) or if you're not using PostgreSQL. ### [django-firebase-mcp](https://mcpcatalogs.com/en/server/raghavdasila-django-firebase-mcp) by raghavdasila · score 39 · ★15 A Django-based MCP server providing 14 Firebase tools for AI agents to interact with Firestore, Auth, and Cloud Storage. **When to choose:** Choose this MCP server if you need AI agents to interact with Firebase services and you're already using Django or want Firebase-specific tools. **When NOT to choose:** Don't choose this if you're not using Firebase or prefer non-Django solutions, as it's tied to Firebase ecosystem and Django framework. ### [mssql-mcp](https://mcpcatalogs.com/en/server/daobataotie-mssql-mcp) by daobataotie · score 39 · ★39 A Python-based MCP server that enables SQL query execution and business intelligence capabilities for MSSQL databases. **When to choose:** Choose this if you need AI-assisted SQL query capabilities specifically for Microsoft SQL Server databases. **When NOT to choose:** Avoid if you need to connect to non-MSSQL databases, as this server is specifically designed for Microsoft SQL Server. ### [mcp-elasticsearch](https://mcpcatalogs.com/en/server/aeazer-mcp-elasticsearch) by AeaZer · score 38 · ★9 A fully-featured Elasticsearch MCP server with comprehensive search capabilities and multiple deployment options. **When to choose:** Choose this server when working with Elasticsearch clusters and need AI agents to perform data operations, search queries, and cluster management through MCP. **When NOT to choose:** Don't choose this if you're using a different search engine or need specialized features not supported by the Elasticsearch API. ### [snowflake-mcp-server](https://mcpcatalogs.com/en/server/dynamike-snowflake-mcp-server) by dynamike · score 36 · ★13 An MCP server for secure read-only queries to Snowflake databases with multiple authentication methods. **When to choose:** Choose this when you need secure, read-only access to Snowflake databases through an MCP interface without modifying any data, especially when using Claude for data analysis. **When NOT to choose:** Avoid if you need write access to Snowflake databases or if you're not already using Snowflake as your data warehouse. ### [SQL_MCP_Server](https://mcpcatalogs.com/en/server/pawankumar94-sql-mcp-server) by pawankumar94 · score 36 · ★13 SQLGenius is an MCP server that converts natural language to SQL queries using Gemini Pro with BigQuery integration. **When to choose:** Choose SQLGenius when you need natural language query access to BigQuery and want a simple UI with visualization capabilities without complex setup. **When NOT to choose:** Don't choose if you need write access to your database, work with non-BigQuery SQL databases, or require authentication beyond the provided service account model. ### [mcp-server-mariadb](https://mcpcatalogs.com/en/server/abel9851-mcp-server-mariadb) by abel9851 · score 34 · ★20 MCP server for read-only access to MariaDB databases through query tools and schema resources. **When to choose:** Choose this when you need AI access to existing MariaDB databases but only require read-only permissions. **When NOT to choose:** Avoid if you need write access to your database or require support for database systems other than MariaDB. ### [mcp-server-unitycatalog](https://mcpcatalogs.com/en/server/ognis1205-mcp-server-unitycatalog) by ognis1205 · score 33 · ★16 Unity Catalog MCP server that provides database functions as tools for AI assistants. **When to choose:** Choose this if you're using Unity Catalog and want to integrate its functions directly into AI applications via MCP. **When NOT to choose:** Don't choose if you need write access beyond function management or if you're not already using Unity Catalog in your data stack. ### [mcp-products-server](https://mcpcatalogs.com/en/server/biobshub-mcp-products-server) by biobshub · score 33 · ★0 A .NET-based MCP server providing product search and marketing summary tools with Cosmos DB integration and a Blazor client UI. **When to choose:** Choose this MCP server if you're working with .NET ecosystems, Azure services, and need product search and marketing content generation capabilities. **When NOT to choose:** Avoid this if you need non-Azure database integration or want to use languages other than C#/.NET for development. ### [ai-agent-dev-week4-mcp-server](https://mcpcatalogs.com/en/server/jmjmlang-ai-agent-dev-week4-mcp-server) by jmjmlang · score 33 · ★0 An MCP server exposing a Person database with CRUD operations for AI agents using Next.js and Prisma. **When to choose:** Choose this MCP server if you need basic CRUD operations for a person database and are already using a Postgres/Neon backend with Next.js. **When NOT to choose:** Don't choose this if you need non-person database operations, custom authentication beyond API keys, or a production-ready solution without workshop-specific dependencies. ### [gahmen-mcp](https://mcpcatalogs.com/en/server/aniruddha-adhikary-gahmen-mcp) by aniruddha-adhikary · score 32 · ★2 MCP server for Singapore government data APIs with search, metadata, and download capabilities. **When to choose:** Choose this server if you need programmatic access to Singapore government datasets and want to search within or download them without building custom API integration. **When NOT to choose:** Don't choose this server if you need access to datasets from countries other than Singapore or require features beyond search and download functionality. ### [kafka-mcp-server](https://mcpcatalogs.com/en/server/joel-hanson-kafka-mcp-server) by Joel-hanson · score 31 · ★1 A MCP server enabling Kafka interaction with tools for topic management and message operations. **When to choose:** Choose this when you need to manage Kafka topics through Claude Desktop and want a simple interface without writing custom Kafka clients. **When NOT to choose:** Don't choose this if you need advanced Kafka operations like message production/consumption, consumer group management, or require robust authentication - these features are not yet implemented. ### [ChatSQL](https://mcpcatalogs.com/en/server/thomasjerard-chatsql) by thomasjerard · score 30 · ★0 ChatSQL converts natural language questions to PostgreSQL queries using Gemini and FastMCP protocol. **When to choose:** Choose ChatSQL if you're using PostgreSQL and want to allow non-technical users to query databases with natural language, with minimal setup. **When NOT to choose:** Avoid ChatSQL if you need support for multiple database types, advanced query features, or if you're concerned about security with sending sensitive database queries to external APIs. ### [redshift-mcp-server](https://mcpcatalogs.com/en/server/amitds1997-redshift-mcp-server) by amitds1997 · score 28 · ★1 A simple MCP server for executing SQL queries on Amazon Redshift clusters with username/password authentication. **When to choose:** Choose this when you need AI to query Redshift data and prefer using username/password authentication instead of more complex authentication methods. **When NOT to choose:** Avoid this if you need write access to Redshift or if you're using a different authentication method like IAM roles. ### [mcp-server](https://mcpcatalogs.com/en/server/carsten-l-mcp-server) by carsten-l · score 26 · ★2 An MCP server that converts text questions about music festivals to database queries and returns plain text responses. **When to choose:** Developers experimenting with MCP protocol and text-to-database query conversion for proof-of-concept implementations. **When NOT to choose:** Production environments need a more mature, documented solution with proper error handling and support. --- ## developer-tools (778) ### [everything](https://mcpcatalogs.com/en/server/modelcontextprotocol-servers-everything) by modelcontextprotocol · score 77 · ★85,748 · official Official MCP test server exercising all protocol features for client builders. **When to choose:** You should choose this when developing or testing MCP clients and need a comprehensive testbed that exercises all protocol features. **When NOT to choose:** Don't choose this for production use cases as it's designed as a test server, not a practical utility. ### [git](https://mcpcatalogs.com/en/server/modelcontextprotocol-servers-git) by modelcontextprotocol · score 75 · ★85,748 · official A comprehensive MCP server for Git operations with status viewing, diffing, commits, and branch management. **When to choose:** Choose this server when you need comprehensive Git operations through an MCP interface, particularly if you're already using TypeScript or prefer a tool with extensive documentation and multiple installation options. **When NOT to choose:** Don't choose this if you need advanced Git features like remote repository management or submodule operations that aren't covered by the current tool set. ### [context7](https://mcpcatalogs.com/en/server/upstash-context7) by upstash · score 65 · ★55,436 Context7 provides up-to-date code documentation and API references through MCP. **When to choose:** Choose Context7 when working with LLM code assistants that need access to up-to-date library documentation, especially when you frequently encounter outdated information or hallucinated API calls. **When NOT to choose:** Avoid Context7 if you need offline access to documentation, as it requires an internet connection to fetch real-time documentation from its service. ### [ruflo](https://mcpcatalogs.com/en/server/ruvnet-ruflo) by ruvnet · score 64 · ★51,795 Multi-agent orchestration platform for Claude Code with MCP server integration, enabling coordinated AI swarms and autonomous workflows. **When to choose:** Choose Ruflo if you need advanced multi-agent orchestration with self-learning capabilities and federated communication across machines. **When NOT to choose:** Don't choose Ruflo if you need a simple, lightweight MCP server for basic tool integration, as it's a comprehensive orchestration platform with significant overhead. ### [fastmcp](https://mcpcatalogs.com/en/server/prefecthq-fastmcp) by PrefectHQ · score 64 · ★25,181 FastMCP is the standard Python framework for building MCP servers, tools, and clients. **When to choose:** Choose FastMCP when building MCP servers in Python, especially if you want automatic schema generation and a simplified API that abstracts away protocol complexities. **When NOT to choose:** Avoid FastMCP if you need non-Python implementation, or require enterprise features like advanced authentication and governance which are provided by their commercial product Horizon. ### [serena](https://mcpcatalogs.com/en/server/oraios-serena) by oraios · score 63 · ★24,287 A powerful MCP toolkit providing semantic code retrieval and editing capabilities for AI agents. **When to choose:** Choose Serena when your AI agents need reliable, IDE-level semantic understanding of codebases for refactoring, navigation, and editing, especially in large or complex multi-language projects. **When NOT to choose:** Don't choose Serena if you only need basic file operations or simple text editing, as it provides advanced semantic capabilities that may be overkill for simple tasks. ### [n8n-mcp](https://mcpcatalogs.com/en/server/czlonkowski-n8n-mcp) by czlonkowski · score 63 · ★20,992 A comprehensive MCP server providing access to 1,650 n8n workflow nodes and 2,352 templates for AI assistants. **When to choose:** Choose n8n-MCP if you're building n8n workflows with AI assistance and need comprehensive knowledge of all available nodes, templates, and configurations. **When NOT to choose:** Don't choose n8n-MCP if you're not using n8n for workflow automation or if you need direct workflow modification capabilities (it's primarily a knowledge provider, not an execution engine). ### [unity-mcp](https://mcpcatalogs.com/en/server/coplaydev-unity-mcp) by CoplayDev · score 61 · ★9,661 Unity MCP connects AI assistants to Unity Editor via Model Context Protocol, enabling direct management of assets, scenes, scripts and automated workflows. **When to choose:** Choose Unity MCP when developing Unity games/applications and want to leverage AI assistants for automation, asset management, and scene editing directly through the MCP protocol. **When NOT to choose:** Not ideal if you're not using Unity Editor, need access to other game engines, or require authentication for secure multi-user environments as the server currently lacks authentication. ### [mcp-use](https://mcpcatalogs.com/en/server/mcp-use-mcp-use) by mcp-use · score 60 · ★9,962 Fullstack MCP framework for building MCP servers and apps that work across ChatGPT, Claude, and other clients. **When to choose:** Choose mcp-use when you need a full-stack solution for building both MCP servers and applications with support for multiple programming languages and ready-to-use templates. **When NOT to choose:** Avoid if you need complete self-hosting without third-party dependencies or if you prefer building MCP servers from scratch without a framework. ### [XcodeBuildMCP](https://mcpcatalogs.com/en/server/getsentry-xcodebuildmcp) by getsentry · score 60 · ★5,594 MCP server providing iOS/macOS development tools for AI agents with CLI interface. **When to choose:** Choose XcodeBuildMCP when working on iOS/macOS development projects and need AI agent assistance with building, testing, and analyzing code within the Xcode ecosystem. **When NOT to choose:** Avoid if you're not working in the Apple ecosystem, as this server is specifically designed for iOS/macOS development and requires Xcode to be installed. ### [fastapi_mcp](https://mcpcatalogs.com/en/server/tadata-org-fastapi-mcp) by tadata-org · score 60 · ★11,863 FastAPI-MCP exposes FastAPI endpoints as MCP tools with built-in authentication. **When to choose:** Choose FastAPI-MCP if you already use FastAPI and want to expose your existing endpoints as MCP tools with minimal configuration and leveraging your existing authentication. **When NOT to choose:** Avoid if you're not using FastAPI, as this solution is specifically designed for FastAPI applications and requires Python 3.10+. ### [csharp-sdk](https://mcpcatalogs.com/en/server/modelcontextprotocol-csharp-sdk) by modelcontextprotocol · score 59 · ★4,264 Official C# SDK for MCP servers and clients with three main packages for different use cases. **When to choose:** Choose this SDK when building MCP servers or clients in .NET environments, especially if you need .NET integration or HTTP-based server capabilities. **When NOT to choose:** Not suitable if you're not using .NET/C# or need MCP functionality in a non-.NET environment. ### [cursor-talk-to-figma-mcp](https://mcpcatalogs.com/en/server/grab-cursor-talk-to-figma-mcp) by grab · score 59 · ★6,761 MCP server connecting AI agents like Cursor and Claude Code with Figma for design automation. **When to choose:** Choose this MCP server if you're a designer or developer using AI tools like Cursor or Claude Code and want to automate Figma design workflows through programmatic access. **When NOT to choose:** Avoid if you need to work with design files outside of Figma, or if your organization doesn't allow third-party plugins in Figma. ### [mcpo](https://mcpcatalogs.com/en/server/open-webui-mcpo) by open-webui · score 58 · ★4,197 A proxy server that exposes MCP tools as OpenAPI-compatible HTTP endpoints for seamless integration with LLM agents. **When to choose:** Choose mcpo when you need to expose MCP tools to systems that only understand OpenAPI, especially when you require security, auto-generated documentation, or OAuth authentication without writing custom integration code. **When NOT to choose:** Avoid mcpo if you're working exclusively within the native MCP ecosystem where stdio communication is sufficient, or if you need low-latency performance as the proxy adds some overhead to MCP calls. ### [ext-apps](https://mcpcatalogs.com/en/server/modelcontextprotocol-ext-apps) by modelcontextprotocol · score 58 · ★2,263 MCP Apps SDK for building interactive UI components that render inline in AI chat clients. **When to choose:** When you need to build rich, interactive UI components that render inline in chat clients beyond simple text responses. **When NOT to choose:** For simple text-only tool responses where UI components are unnecessary, or if you need client-specific features not covered by the standard. ### [codebase-memory-mcp](https://mcpcatalogs.com/en/server/deusdata-codebase-memory-mcp) by DeusData · score 58 · ★2,363 High-performance code intelligence MCP server that indexes codebases into a persistent knowledge graph with sub-ms queries and 155 language support. **When to choose:** Choose this for large codebases that need fast, structural analysis across multiple languages with local processing and minimal token usage. **When NOT to choose:** Don't choose if you need write access to the codebase, as this is a read-only indexing tool, or if you prefer cloud-based solutions. ### [shadcn-ui-mcp-server](https://mcpcatalogs.com/en/server/jpisnice-shadcn-ui-mcp-server) by Jpisnice · score 58 · ★2,768 MCP server providing AI assistants with comprehensive access to shadcn/ui components for React, Svelte, Vue, and React Native. **When to choose:** Choose this server when working with shadcn/ui components across multiple frameworks and need AI assistance with implementation details. **When NOT to choose:** Not ideal if you only need components for a single framework or require authentication beyond GitHub token access. ### [xiaozhi-esp32-server](https://mcpcatalogs.com/en/server/xinnan-tech-xiaozhi-esp32-server) by xinnan-tech · score 58 · ★9,554 MCP server for ESP32 device management with voice recognition, MQTT, and multi-language support. **When to choose:** Choose this MCP server when building voice-controlled IoT systems with ESP32 devices and need integrated AI capabilities like speech recognition and multi-modal interactions. **When NOT to choose:** Don't choose this if you're not using ESP32 hardware, as it's tightly coupled to this platform. Also consider alternatives if you need a more generalized MCP implementation without extensive voice/IoT features. ### [klavis](https://mcpcatalogs.com/en/server/klavis-ai-klavis) by Klavis-AI · score 58 · ★5,735 Klavis is an MCP integration platform offering over 100 prebuilt connectors with OAuth support for AI agents. **When to choose:** Choose Klavis when you need scalable MCP integration for multiple services and want ready-to-use connectors with OAuth support. **When NOT to choose:** Avoid Klavis if you need complete open-source control without vendor lock-in, or if you're looking for a single self-contained MCP server without cloud dependencies. ### [design-extract](https://mcpcatalogs.com/en/server/manavarya09-design-extract) by Manavarya09 · score 57 · ★2,640 A comprehensive MCP server that extracts design systems from websites, generating tokens, components, and multi-platform outputs. **When to choose:** Choose design-extract when you need to reverse-engineer design systems from live websites for implementation in other projects, especially when you need multi-platform code generation or design system analysis capabilities. **When NOT to choose:** Avoid if you need to analyze non-web design systems like desktop applications or if you require write access to modify the source design system (this tool is read-only). ### [mcphub](https://mcpcatalogs.com/en/server/samanhappy-mcphub) by samanhappy · score 57 · ★2,082 MCPHub is a unified management platform for orchestrating multiple MCP servers with flexible routing and centralized dashboard. **When to choose:** Choose MCPHub when you need to manage multiple MCP servers in production with enterprise features like authentication, routing controls, and monitoring. **When NOT to choose:** Avoid if you only need to run a single MCP server or want a lightweight solution without dashboard complexity. ### [lean-ctx](https://mcpcatalogs.com/en/server/yvgude-lean-ctx) by yvgude · score 57 · ★1,676 lean-ctx is an MCP server that reduces token waste in AI coding tools by 60-95% through intelligent context compression and caching. **When to choose:** Choose lean-ctx if you regularly use AI coding assistants in medium-to-large repositories and want to significantly reduce token costs while maintaining full context awareness. **When NOT to choose:** Avoid lean-ctx if you primarily work with very small repositories, don't use shell commands from your AI tools, or need raw/unfiltered logs for debugging. ### [Unity-MCP](https://mcpcatalogs.com/en/server/ivanmurzak-unity-mcp) by IvanMurzak · score 57 · ★2,767 Unity-MCP enables AI agents to work directly in Unity Editor and runtime through MCP protocol. **When to choose:** Choose Unity MCP if you're developing with Unity and want to integrate AI directly into your development workflow or game runtime. **When NOT to choose:** Don't choose Unity MCP if you're not using Unity or need AI integration with other game engines. ### [inspector](https://mcpcatalogs.com/en/server/mcpjam-inspector) by MCPJam · score 56 · ★1,944 MCPJam is a comprehensive development platform for debugging, inspecting, and evaluating MCP servers and apps. **When to choose:** Choose MCPJam Inspector when you need a comprehensive debugging platform for MCP servers with advanced features like OAuth testing, multi-model evaluation, and CI/CD integration. **When NOT to choose:** Avoid MCPJam Inspector if you need a simple lightweight tool for basic MCP server inspection or if you prefer open-source self-hosted solutions over commercial offerings. ### [fast-agent](https://mcpcatalogs.com/en/server/evalstate-fast-agent) by evalstate · score 56 · ★3,779 A comprehensive MCP server framework for building and evaluating AI agents with tool calling support. **When to choose:** Choose fast-agent when you need a comprehensive MCP-first framework for building sophisticated agents with workflow orchestration, especially if you require multiple model providers and advanced diagnostics. **When NOT to choose:** Avoid fast-agent if you need a pure GUI-based development environment or if you're working with systems that require extensive authentication customization beyond OAuth. ### [contextplus](https://mcpcatalogs.com/en/server/forloopcodes-contextplus) by forloopcodes · score 56 · ★1,896 Context+ is an MCP server that transforms codebases into searchable feature graphs using RAG, AST parsing, and semantic linking. **When to choose:** Choose Context+ when working on large, complex codebases requiring deep semantic understanding, dependency analysis, and knowledge graph capabilities beyond simple code search. **When NOT to choose:** Avoid Context+ for small projects or when you need lightweight, simple code navigation without the overhead of embedding models and graph storage requirements. ### [apple-docs-mcp](https://mcpcatalogs.com/en/server/kimsungwhee-apple-docs-mcp) by kimsungwhee · score 56 · ★1,259 Apple Developer Documentation MCP server providing searchable access to official docs, APIs, WWDC videos, and code examples across Apple platforms. **When to choose:** Choose this MCP server when you're an Apple developer working with iOS, macOS, or other Apple platforms and want instant access to official documentation, APIs, and WWDC content within your AI assistant. **When NOT to choose:** Don't choose this if you're not developing for Apple platforms, as it provides no value for non-Apple documentation or frameworks. ### [mcp-unity](https://mcpcatalogs.com/en/server/codergamester-mcp-unity) by CoderGamester · score 56 · ★1,699 MCP server for Unity Editor enabling AI assistants to manipulate scenes, GameObjects, and other Unity assets. **When to choose:** Choose MCP Unity if you're a Unity developer who wants to use AI assistants like Cursor or Claude Code to directly manipulate Unity projects and scenes through natural language. **When NOT to choose:** Don't choose MCP Unity if you're using an older Unity version (pre-6), need extensive write operations beyond Unity manipulation, or require a more general-purpose game development assistant. ### [mcp-windbg](https://mcpcatalogs.com/en/server/svnscha-mcp-windbg) by svnscha · score 55 · ★1,288 MCP server that bridges AI models with WinDbg for Windows crash dump analysis and remote debugging. **When to choose:** Choose this server when you need AI assistance analyzing Windows crash dumps or performing live debugging sessions with WinDbg/CDB through natural language commands. **When NOT to choose:** Don't choose this if you need to debug non-Windows systems or if you're looking for an automatic crash fixing solution rather than an AI-assisted debugging tool. ### [pi-mcp-adapter](https://mcpcatalogs.com/en/server/nicobailon-pi-mcp-adapter) by nicobailon · score 55 · ★689 Token-efficient MCP adapter for Pi coding agent that reduces context window usage by lazy-loading MCP servers. **When to choose:** Choose this when using Pi with multiple MCP servers and you're hitting context limits, or when you want to integrate existing MCP configurations without configuration overhead. **When NOT to choose:** Don't choose this if you're not using Pi, or if you prefer all MCP tools to be directly available without proxy discovery. ### [mcp-language-server](https://mcpcatalogs.com/en/server/isaacphi-mcp-language-server) by isaacphi · score 55 · ★1,527 MCP language server bridges LSP tools to MCP clients enabling code navigation, definitions, references and diagnostics. **When to choose:** Choose this when you need to connect LLM clients to existing language servers for code analysis and navigation across multiple programming languages. **When NOT to choose:** Avoid this if you need advanced IDE features beyond basic language server capabilities or if you're working with a language server that doesn't communicate over stdio. ### [arcade-mcp](https://mcpcatalogs.com/en/server/arcadeai-arcade-mcp) by ArcadeAI · score 55 · ★892 Python framework for building MCP servers with OAuth, tool development, and vendor-neutral integration. **When to choose:** Choose arcade-mcp when you need to build custom MCP servers with OAuth authentication or integrate with specific APIs not available in prebuilt catalogs, especially if you want vendor neutrality across different LLM clients. **When NOT to choose:** Avoid arcade-mcp if you need a framework for languages other than Python, or if you prefer self-hosted solutions without dependency on Arcade Cloud for advanced features like token management. ### [metamcp](https://mcpcatalogs.com/en/server/metatool-ai-metamcp) by metatool-ai · score 55 · ★2,328 MetaMCP is a unified MCP proxy that aggregates multiple MCP servers into one with middleware support. **When to choose:** Choose MetaMCP when you need to manage multiple MCP servers in a unified infrastructure with authentication, tool overrides, and middleware capabilities. **When NOT to choose:** Avoid MetaMCP for simple use cases with just one MCP server or if you require direct tool access without aggregation layer overhead. ### [mcphub.nvim](https://mcpcatalogs.com/en/server/ravitemer-mcphub-nvim) by ravitemer · score 55 · ★1,765 Neovim client that integrates MCP servers with chat plugins for enhanced development workflows. **When to choose:** Choose this if you're a Neovim user looking to integrate MCP servers into your editing workflow with an intuitive management interface. **When NOT to choose:** Don't choose this if you're not using Neovim or if you need to create MCP servers rather than consume them. ### [mcp-router](https://mcpcatalogs.com/en/server/mcp-router-mcp-router) by mcp-router · score 55 · ★2,011 MCP Router is a desktop application for managing multiple MCP servers with centralized control. **When to choose:** Choose this if you're managing multiple MCP servers and need a centralized dashboard to organize, toggle, and monitor them. **When NOT to choose:** Don't choose this if you're looking for an MCP server itself rather than a management tool for MCP servers. ### [chunkhound](https://mcpcatalogs.com/en/server/chunkhound-chunkhound) by chunkhound · score 55 · ★1,258 Local-first codebase intelligence tool with MCP integration for semantic code searching across 32+ languages. **When to choose:** Choose ChunkHound when you need deep code relationship analysis in security-sensitive environments or large multi-language codebases where traditional search falls short. **When NOT to choose:** Avoid ChunkHound if you need cloud-based search with minimal local setup or if you're working exclusively with codebases that don't benefit from semantic relationship analysis. ### [ros-mcp-server](https://mcpcatalogs.com/en/server/robotmcp-ros-mcp-server) by robotmcp · score 55 · ★1,224 ROS MCP Server connects AI models like Claude & GPT with robots via ROS without changing existing robot code. **When to choose:** Choose this when you need AI models to interact with robots running ROS without modifying existing robot code. **When NOT to choose:** Don't choose this if your robot doesn't use ROS, or if you need tight integration with non-MCP AI platforms. ### [golf](https://mcpcatalogs.com/en/server/golf-mcp-golf) by golf-mcp · score 55 · ★824 Production-ready MCP server framework for building, deploying and scaling secure AI agent infrastructure with authentication, observability, and debugging capabilities. **When to choose:** Choose Golf when building production-ready MCP servers that require enterprise-grade authentication, observability, and a clean Python-based development experience. **When NOT to choose:** Avoid Golf if you need a lightweight solution without authentication, prefer non-Python implementations, or require minimal framework dependencies. ### [mcptools](https://mcpcatalogs.com/en/server/f-mcptools) by f · score 55 · ★1,584 A comprehensive CLI for MCP servers with tools discovery, mock/proxy functionality, and web interface. **When to choose:** Choose MCP Tools when you need a comprehensive CLI for managing MCP servers across different transports, creating mock environments for testing, or building proxy integrations with shell scripts. **When NOT to choose:** Don't choose MCP Tools if you need a simple MCP client for basic tool interactions or prefer a GUI interface for server management. ### [mcp-proxy](https://mcpcatalogs.com/en/server/sparfenyuk-mcp-proxy) by sparfenyuk · score 55 · ★2,520 MCP proxy that bridges stdio and SSE/StreamableHTTP transports for connecting MCP servers. **When to choose:** Choose this when you need to connect MCP clients that only support stdio transport with remote MCP servers using SSE/StreamableHTTP, or when you want to expose local MCP servers to remote clients. **When NOT to choose:** Avoid if you need direct communication between stdio-only MCP servers and clients without the proxy layer, or if you require WebSocket transport which is not supported. ### [cupertino](https://mcpcatalogs.com/en/server/mihaelamj-cupertino) by mihaelamj · score 55 · ★742 A Swift-based MCP server that indexes and serves Apple's developer documentation to AI agents. **When to choose:** Choose Cupertino when working with Apple platform development and you need accurate, offline access to Apple's official documentation through AI assistants. **When NOT to choose:** Don't choose Cupertino if you need cross-platform support (macOS-only), have very limited disk space, or primarily work with non-Apple documentation. ### [antigravity-workspace-template](https://mcpcatalogs.com/en/server/study8677-antigravity-workspace-template) by study8677 · score 55 · ★1,219 Multi-agent knowledge engine that turns any codebase into a queryable AI assistant with MCP integration. **When to choose:** Choose Antigravity when working with complex codebases that need AI-powered code exploration and documentation generation, particularly if you're already using Claude Code or Codex CLI. **When NOT to choose:** Avoid Antigravity if you need write access to code (this is read-only), require offline functionality, or prefer a more traditional documentation approach over AI-generated knowledge. ### [next-devtools-mcp](https://mcpcatalogs.com/en/server/vercel-next-devtools-mcp) by vercel · score 54 · ★747 Next.js DevTools MCP server providing development tools and utilities for Next.js applications. **When to choose:** Choose this MCP server when working with Next.js projects and need runtime diagnostics, automatic documentation access, or development automation within AI coding assistants. **When NOT to choose:** Avoid if you're not using Next.js, or prefer a more general-purpose web development MCP server without framework-specific optimizations. ### [semble](https://mcpcatalogs.com/en/server/minishlab-semble) by MinishLab · score 54 · ★823 High-performance code search library that runs as an MCP server with 98% fewer tokens than grep+read. **When to choose:** Choose Semble when you need fast, accurate code search for AI agents with minimal token usage, especially for large codebases where traditional grep approaches are inefficient. **When NOT to choose:** Don't choose Semble if you need exhaustive literal text searches, require very recent features from a project with only a few months of active development, or need write access to code (it's read-only). ### [ref-tools-mcp](https://mcpcatalogs.com/en/server/ref-tools-ref-tools-mcp) by ref-tools · score 54 · ★1,104 Ref MCP provides token-efficient API documentation search tools to reduce context rot and minimize costs. **When to choose:** Choose Ref MCP when your coding agent needs efficient access to documentation and you're willing to use a cloud service that minimizes token usage and context rot. **When NOT to choose:** Don't choose Ref if you need offline-only access, have concerns about vendor lock-in, or require handling of highly sensitive documentation that can't be shared with external services. ### [mcp-framework](https://mcpcatalogs.com/en/server/quantgeekdev-mcp-framework) by QuantGeekDev · score 54 · ★916 TypeScript framework for building MCP servers with automatic discovery, type safety, and multiple transport options. **When to choose:** Choose this framework when building MCP servers in TypeScript, especially when you need multiple transport protocols, built-in authentication, and a comprehensive CLI for project management. **When NOT to choose:** Don't choose this framework if you're building a simple MCP server with basic requirements or if you prefer using vanilla JavaScript without TypeScript. ### [cheatengine-mcp-bridge](https://mcpcatalogs.com/en/server/miscusi-peek-cheatengine-mcp-bridge) by miscusi-peek · score 54 · ★691 MCP server connecting AI assistants directly to Cheat Engine for automated reverse engineering and memory analysis. **When to choose:** When performing reverse engineering, game modding, or security audits where you need to automate memory analysis and pointer scanning using AI assistance. **When NOT to choose:** If you're not working on Windows, need to analyze non-x86 architectures, or require stable, long-term support as this is a specialized tool with a narrow focus. ### [MCP-Nest](https://mcpcatalogs.com/en/server/rekog-labs-mcp-nest) by rekog-labs · score 54 · ★647 A NestJS module for creating MCP servers with tools, resources, and prompts using dependency injection. **When to choose:** Choose MCP-Nest when you're already using NestJS for your backend and want to build MCP servers with enterprise-grade authentication and tooling. **When NOT to choose:** Avoid if you're not using TypeScript/NestJS or need a more lightweight solution without authentication complexity. ### [cclsp](https://mcpcatalogs.com/en/server/ktnyt-cclsp) by ktnyt · score 54 · ★637 cclsp is an MCP server that bridges AI coding assistants with LSP servers for robust code navigation and symbol manipulation. **When to choose:** Choose cclsp when you need AI assistants to navigate and analyze codebases with LSP capabilities, especially when working with multiple programming languages in non-IDE environments. **When NOT to choose:** Don't choose cclsp if you're already satisfied with your IDE's built-in AI integration, or if you're working exclusively in environments that already have robust LSP support without AI augmentation. ### [roam-code](https://mcpcatalogs.com/en/server/cranot-roam-code) by Cranot · score 54 · ★461 A local codebase intelligence CLI and MCP server that analyzes code structures, dependencies, and patterns to provide actionable insights for developers and AI agents. **When to choose:** Choose roam-code when working with AI coding assistants on complex codebases where you need to understand structural relationships and ensure algorithmic correctness beyond traditional code review. **When NOT to choose:** Avoid roam-code if you need a simple, lightweight linter or if your workflow requires cloud-based analysis since it runs entirely locally with no SaaS offering. ### [hyper-mcp](https://mcpcatalogs.com/en/server/hyper-mcp-rs-hyper-mcp) by hyper-mcp-rs · score 54 · ★871 A fast, secure MCP server with WebAssembly plugin support for extending AI capabilities. **When to choose:** Choose hyper-mcp when you need a fast, secure way to extend AI capabilities with plugins written in multiple languages and deployed across various environments from cloud to edge. **When NOT to choose:** Avoid if you need plugins that require access to system resources beyond what WebAssembly sandboxing allows, or if you're working in an environment without cosign available for OCI plugin verification. ### [mcp-nixos](https://mcpcatalogs.com/en/server/utensils-mcp-nixos) by utensils · score 54 · ★639 MCP-NixOS provides real-time NixOS package, option, and documentation data to prevent AI hallucinations. **When to choose:** Choose MCP-NixOS when your AI system needs accurate, up-to-date information about Nix packages, options, and configuration without hallucination. **When NOT to choose:** Don't choose this if you need access to non-Nix systems or require write access to package repositories - this is a read-only data source. ### [fusio](https://mcpcatalogs.com/en/server/apioo-fusio) by apioo · score 54 · ★2,088 Fusio is a self-hosted API management platform with native MCP support to expose APIs as tools for AI agents. **When to choose:** Choose Fusio if you need to transform existing APIs or databases into MCP tools for AI agents and want a self-hosted solution with API monetization capabilities. **When NOT to choose:** Avoid Fusio if you need a fully-managed cloud solution or if you prefer a simpler, direct MCP integration without the overhead of a full API management platform. ### [MCPJungle](https://mcpcatalogs.com/en/server/mcpjungle-mcpjungle) by mcpjungle · score 54 · ★1,041 A self-hosted MCP gateway that centralizes multiple MCP servers behind one endpoint. **When to choose:** Choose MCPJungle when you need to manage multiple MCP servers and want a single endpoint for all your AI clients. **When NOT to choose:** Avoid if you only use a single MCP server, as MCPJungle adds complexity to a simple setup. ### [jupyter-mcp-server](https://mcpcatalogs.com/en/server/datalayer-jupyter-mcp-server) by datalayer · score 54 · ★1,103 An MCP server that enables AI systems to interact with Jupyter notebooks in real-time, providing tools for notebook management and cell execution. **When to choose:** Choose Jupyter MCP Server if you need AI to interact with Jupyter notebooks in real-time, especially if you're already using Jupyter for data analysis or machine learning workflows. **When NOT to choose:** Avoid if you need highly customized Jupyter extensions beyond what's provided, or if you're not using Jupyter notebooks for your workflow. ### [shippie](https://mcpcatalogs.com/en/server/mattzcarey-shippie) by mattzcarey · score 54 · ★2,356 Shippie is an MCP client for code review that integrates AI models into CI/CD pipelines. **When to choose:** Choose Shippie when you need AI-powered code review integrated into your CI/CD pipeline and want to extend functionality via MCP client connections. **When NOT to choose:** Don't choose Shippie if you need an MCP server rather than a client, or if you prefer human-only code review processes without AI assistance. ### [wcgw](https://mcpcatalogs.com/en/server/rusiaaman-wcgw) by rusiaaman · score 54 · ★660 An MCP server providing shell and code editing tools with interactive terminal access for AI agents. **When to choose:** Choose wcgw when you need a highly interactive shell experience where both human and AI can control the same terminal, with advanced file editing capabilities and task checkpointing. **When NOT to choose:** Avoid wcgw if you need strict security boundaries for AI access, or if you're working in an environment where unfiltered shell access poses unacceptable risks. ### [memorix](https://mcpcatalogs.com/en/server/avids2-memorix) by AVIDS2 · score 53 · ★449 Open-source cross-agent memory layer for coding agents via MCP with local-first storage. **When to choose:** Choose Memorix when you need persistent memory across multiple IDEs and coding agents, especially if you work with multiple AI coding tools and want them to share project context and reasoning. **When NOT to choose:** Don't choose Memorix if you only use one IDE and coding agent, prefer cloud-based solutions over local-first storage, or need a simpler memory system without the complexity of multi-agent orchestration. ### [concierge](https://mcpcatalogs.com/en/server/concierge-hq-concierge) by concierge-hq · score 53 · ★534 Concierge is a Python SDK that enhances MCP servers with progressive tool disclosure, workflow state management, and semantic search capabilities. **When to choose:** Choose Concierge when building complex MCP servers with many tools that need progressive disclosure, enforced workflows, or semantic search capabilities. **When NOT to choose:** Avoid Concierge if you need a simple MCP server with minimal tools or if you prefer working directly with the MCP protocol without the abstraction layer. ### [GhidrAssistMCP](https://mcpcatalogs.com/en/server/symgraph-ghidrassistmcp) by symgraph · score 53 · ★613 A Ghidra extension providing MCP server functionality for AI assistants to interact with reverse engineering capabilities. **When to choose:** Choose this when you need AI assistance for reverse engineering tasks with Ghidra and want seamless integration between AI models and Ghidra's analysis capabilities. **When NOT to choose:** Don't choose this if you're not using Ghidra for reverse engineering, as it's specifically designed as a Ghidra extension. ### [codanna](https://mcpcatalogs.com/en/server/bartolli-codanna) by bartolli · score 53 · ★673 Codanna is a local code intelligence MCP server providing semantic search, relationship tracking, and document search for AI coding assistants. **When to choose:** Choose Codanna when you need fast, semantic code intelligence for AI agents in large codebases and LSP is too slow. **When NOT to choose:** Don't choose Codanna if you need write access to code, real-time collaboration features, or require Windows support which is still experimental. ### [mcp](https://mcpcatalogs.com/en/server/cloudflare-mcp) by cloudflare · score 53 · ★459 Token-efficient MCP server for Cloudflare API with 2500 endpoints accessible via Code Mode. **When to choose:** Choose this server when you need to interact with Cloudflare's extensive API ecosystem through an AI agent and want maximum token efficiency. **When NOT to choose:** Avoid if you need fine-grained control over each API endpoint as an individual tool (requires disabling Code Mode) or if you're not already using Cloudflare services. ### [reverse-engineering-assistant](https://mcpcatalogs.com/en/server/cyberkaida-reverse-engineering-assistant) by cyberkaida · score 53 · ★731 Ghidra MCP server for AI-powered reverse engineering with specialized tools for binary analysis. **When to choose:** Choose ReVa when you need AI assistance for reverse engineering tasks and already use or are willing to use Ghidra as your primary RE tool. **When NOT to choose:** Don't choose ReVa if you need support for other reverse engineering tools like IDA Pro or Binary Ninja, or if you prefer a GUI-based AI assistant over tool-driven approaches. ### [matlab-mcp-core-server](https://mcpcatalogs.com/en/server/matlab-matlab-mcp-core-server) by matlab · score 53 · ★636 Official MATLAB MCP Server enabling AI applications to execute MATLAB code and access MATLAB functionality. **When to choose:** Choose this when working with MATLAB environments and need AI assistance for code development, analysis, or execution within MATLAB ecosystem. **When NOT to choose:** Avoid if you don't have a MATLAB license as this server requires a valid MATLAB installation, and consider alternatives if you're concerned about vendor lock-in with MathWorks. ### [idea-reality-mcp](https://mcpcatalogs.com/en/server/mnemox-ai-idea-reality-mcp) by mnemox-ai · score 53 · ★691 MCP server that scans GitHub, HN, npm, PyPI, Product Hunt to check if an app idea already exists. **When to choose:** Choose this when building a new project and want to automatically check for existing competitors and market trends. **When NOT to choose:** Don't choose this for non-software ideas or when you need more granular market analysis beyond basic competition checks. ### [DebugMCP](https://mcpcatalogs.com/en/server/microsoft-debugmcp) by microsoft · score 53 · ★349 DebugMCP enables AI agents to debug code in VS Code with breakpoints, stepping, and variable inspection. **When to choose:** Choose DebugMCP when you need AI agents to perform real debugging operations in VS Code, including breakpoints, stepping through code, and inspecting variables. **When NOT to choose:** Avoid if you need debugging capabilities outside of VS Code or require integration with non-MCP-based AI assistants. ### [obot](https://mcpcatalogs.com/en/server/obot-platform-obot) by obot-platform · score 53 · ★777 Complete MCP platform offering hosting, registry, gateway, and chat client. **When to choose:** Choose Obot when you need a comprehensive solution for hosting, managing, and discovering MCP servers across your organization with built-in security and monitoring. **When NOT to choose:** Don't choose Obot if you simply need a single MCP server for a specific task, as it's a full platform that requires significant resources to deploy and maintain. ### [Unreal_mcp](https://mcpcatalogs.com/en/server/chir24-unreal-mcp) by ChiR24 · score 53 · ★622 MCP server enabling AI assistants to control Unreal Engine through native C++ Automation Bridge plugin. **When to choose:** Choose this when you need AI-powered control of Unreal Engine workflows, especially for game development, virtual production, or automated testing scenarios. **When NOT to choose:** Avoid this if you're not using Unreal Engine, need access to other game engines, or require a solution that works without compiling native plugins. ### [claude-talk-to-figma-mcp](https://mcpcatalogs.com/en/server/arinspunk-claude-talk-to-figma-mcp) by arinspunk · score 53 · ★593 An MCP server that enables AI agents to interact with Figma designs through reading, analysis, and modification capabilities. **When to choose:** Choose this MCP server when you need AI agents to interact with Figma designs and you don't have a paid Figma Dev Mode license. **When NOT to choose:** Don't choose this if you need advanced Figma Variables support or are working exclusively in a team environment with strict Figma Dev Mode licensing. ### [sonarqube-mcp-server](https://mcpcatalogs.com/en/server/sonarsource-sonarqube-mcp-server) by SonarSource · score 53 · ★554 SonarQube MCP Server enables code quality and security analysis within AI agents via the MCP protocol. **When to choose:** Choose this when you need to integrate code quality analysis into AI workflows and are already using SonarQube. **When NOT to choose:** Don't choose if you're not using SonarQube or need to integrate with other code quality tools like SonarCloud or static analysis platforms. ### [clawdcursor](https://mcpcatalogs.com/en/server/amrdab-clawdcursor) by AmrDab · score 53 · ★315 An OS-agnostic, model-agnostic desktop automation server that gives AI agents eyes, hands, and verification through MCP. **When to choose:** Choose Clawd Cursor when you need to automate GUI interactions where APIs don't exist, especially with legacy applications, Electron apps, or desktop software behind authentication. **When NOT to choose:** Avoid Clawd Cursor when working with applications that already have reliable APIs, when performance is critical, or when the automation task can be handled more efficiently via CLI commands or direct file manipulation. ### [mcp-for-argocd](https://mcpcatalogs.com/en/server/argoproj-labs-mcp-for-argocd) by argoproj-labs · score 53 · ★464 MCP server enabling AI assistants to interact with Argo CD applications through natural language. **When to choose:** Choose this server if you're using Argo CD for Kubernetes application delivery and want AI assistants to manage applications through natural language commands. **When NOT to choose:** Don't choose this if you need access to other GitOps tools like Flux or if your Argo CD instance lacks API access tokens for authentication. ### [Overture](https://mcpcatalogs.com/en/server/sixhq-overture) by SixHq · score 53 · ★621 Overture is an MCP server that visualizes AI coding agent execution plans as interactive flowcharts before code is written. **When to choose:** Choose Overture when you need visibility into and control over AI coding agent workflows, especially for complex projects where understanding the execution plan before code generation is valuable. **When NOT to choose:** Avoid Overture if you prefer direct AI coding without planning visualization, or if you're working with a non-MCP compatible AI agent that doesn't support the MCP protocol. ### [docfork](https://mcpcatalogs.com/en/server/docfork-docfork) by docfork · score 53 · ★479 Docfork provides AI coding agents with real-time documentation access through MCP servers to prevent API hallucinations. **When to choose:** Choose Docfork when working with multiple libraries and frameworks where up-to-date documentation is critical for accurate AI-generated code. **When NOT to choose:** Don't choose Docfork if you need offline access to documentation or prefer a self-hosted solution without external API dependencies. ### [sentry-mcp](https://mcpcatalogs.com/en/server/getsentry-sentry-mcp) by getsentry · score 53 · ★691 MCP server for Sentry integration with LLMs, providing error tracking and debugging capabilities. **When to choose:** Choose this server when you're already using Sentry for error tracking and want to integrate it directly with AI coding assistants for debugging and issue resolution. **When NOT to choose:** Don't choose this if you're not using Sentry or need read/write access to your error tracking system beyond what Sentry provides. ### [tda](https://mcpcatalogs.com/en/server/irockel-tda) by irockel · score 53 · ★542 TDA is a Java thread dump analyzer with MCP server capabilities for AI-powered diagnostics. **When to choose:** Choose TDA when you need AI-powered analysis of Java thread dumps, especially when working with modern Java applications using virtual threads or when integrating with AI development tools like Cursor or Junie. **When NOT to choose:** Don't choose TDA if you need analysis for non-Java applications or if you require extensive visualization capabilities beyond what the GUI or MCP server provides. ### [xmcp](https://mcpcatalogs.com/en/server/basementstudio-xmcp) by basementstudio · score 53 · ★1,252 A TypeScript framework for building MCP servers with file system routing, hot reloading, and extensible configuration. **When to choose:** Choose xmcp when building TypeScript-based MCP servers that need file system routing, hot reloading, and middleware support for authentication. **When NOT to choose:** Avoid xmcp if you need a solution for non-TypeScript environments or if you prefer more minimal frameworks without built-in features like file system routing. ### [home-assistant-vibecode-agent](https://mcpcatalogs.com/en/server/coolver-home-assistant-vibecode-agent) by Coolver · score 53 · ★550 An MCP server that enables AI IDEs to manage Home Assistant through natural language for automation creation, dashboard design, and configuration management. **When to choose:** Choose this if you want to use AI assistants like Claude Code or Cursor to manage your Home Assistant automations and configurations without manual YAML editing. **When NOT to choose:** Don't choose this if you're new to Home Assistant (requires experience), need add-on management but use a container without Supervisor, or want a solution without automatic versioning. ### [flutter-skill](https://mcpcatalogs.com/en/server/ai-dashboad-flutter-skill) by ai-dashboad · score 52 · ★252 An AI-powered E2E testing tool for 10 platforms with 253 MCP tools, enabling testing from natural language without writing test code. **When to choose:** Choose flutter-skill when you need AI-powered, cross-platform E2E testing without writing test code or maintaining selectors, especially for Flutter or React Native applications. **When NOT to choose:** Don't choose flutter-skill if you need fine-grained control over test automation, require traditional test frameworks like Jest/Cypress, or need to test applications that cannot be modified to include the flutter-skill SDK. ### [godot-mcp-pro](https://mcpcatalogs.com/en/server/youichi-uda-godot-mcp-pro) by youichi-uda · score 52 · ★340 Premium MCP server providing 172 tools for AI-powered Godot 4 game development with real-time editor integration. **When to choose:** Godot developers who want direct AI interaction with their game engine for automated development, testing, and creative assistance through a premium, well-maintained MCP server. **When NOT to choose:** Budget-conscious developers or those who prefer open-source solutions, as this is a premium product requiring a one-time payment with no free tier. ### [mcp-proxy](https://mcpcatalogs.com/en/server/tbxark-mcp-proxy) by tbxark · score 52 · ★691 MCP proxy server that aggregates multiple MCP servers behind a single HTTP entrypoint with flexible configuration options. **When to choose:** Choose this proxy when managing multiple MCP servers and you need a single entry point to access all their resources, especially if you're already using Go-based infrastructure or Docker deployments. **When NOT to choose:** Avoid this if you need fine-grained control over individual MCP servers or if you're working in an environment where Go-based services are not preferred. ### [stackql](https://mcpcatalogs.com/en/server/stackql-stackql) by stackql · score 52 · ★841 StackQL is a Go-based SQL framework for querying and managing cloud resources and APIs via MCP protocol. **When to choose:** Choose StackQL when you need to query and manage multiple cloud resources using SQL syntax and want a consistent interface across different providers. **When NOT to choose:** Don't choose StackQL if you need real-time operational capabilities or complex transactional support, as it focuses on read-heavy querying and provisioning tasks. ### [Context-Engine](https://mcpcatalogs.com/en/server/context-engine-ai-context-engine) by Context-Engine-AI · score 52 · ★392 Context Engine MCP provides semantic search, code navigation, and memory tools for AI coding assistants. **When to choose:** Choose Context Engine when you need advanced semantic code search and symbol intelligence features for AI coding assistants, especially working with large codebases that benefit from batch queries and cross-repo tracing. **When NOT to choose:** Avoid this if you need a fully open-source solution (uses proprietary SDE license), or if you're looking for a lightweight tool without the dependency on external authentication. ### [open-codex-computer-use](https://mcpcatalogs.com/en/server/ifuryst-open-codex-computer-use) by iFurySt · score 52 · ★801 Open-source MCP server for Computer Use on macOS, Linux, and Windows, inspired by Codex Computer Use. **When to choose:** Choose this when you need cross-platform computer automation capabilities for your AI agent and prefer an open-source solution over OpenAI's Codex Computer Use. **When NOT to choose:** Not suitable for Windows and Linux users requiring advanced accessibility features, as the project's primary focus and optimization is for macOS. ### [elementor-mcp](https://mcpcatalogs.com/en/server/msrbuilds-elementor-mcp) by msrbuilds · score 52 · ★318 WordPress plugin that extends Elementor with 97 MCP tools for AI-powered page design management. **When to choose:** Choose this MCP server if you're working with WordPress and Elementor and need programmatic control over page designs through AI agents. **When NOT to choose:** Not suitable if you're not using WordPress or Elementor, or if you need more direct access to the database. ### [git-mcp-server](https://mcpcatalogs.com/en/server/cyanheads-git-mcp-server) by cyanheads · score 52 · ★215 A comprehensive MCP server that provides AI agents with 28 Git operations including clone, commit, branch management, and more. **When to choose:** Choose this when you need comprehensive Git operations through MCP for AI agents, especially when you're already working with TypeScript-based systems and need both STDIO and HTTP support. **When NOT to choose:** Avoid this if you need a pure JavaScript solution without TypeScript dependencies or require direct integration with specific Git providers beyond standard CLI. ### [octocode-mcp](https://mcpcatalogs.com/en/server/bgauryy-octocode-mcp) by bgauryy · score 52 · ★832 MCP server for semantic code research across GitHub/GitLab and local repositories with advanced LSP integration. **When to choose:** Choose this if you're already using GitHub/GitLab and need AI-powered code search and analysis with context-aware development assistance. **When NOT to choose:** Don't choose this if you need to work with code outside GitHub/GitLab or if you require a standalone code analysis solution without cloud dependencies. ### [vurb.ts](https://mcpcatalogs.com/en/server/vinkius-labs-vurb-ts) by vinkius-labs · score 52 · ★251 Production-grade TypeScript framework for building MCP servers with presenters, FSM gating, and semantic routing. **When to choose:** Choose Vurb.ts when you need production-ready MCP servers with governance controls, PII redaction, and state-based tool gating. **When NOT to choose:** Don't choose Vurb.ts if you need a lightweight solution without complex governance or if you prefer raw MCP implementations with minimal abstraction. ### [bernstein](https://mcpcatalogs.com/en/server/sipyourdrink-ltd-bernstein) by sipyourdrink-ltd · score 52 · ★375 Audit-grade multi-agent orchestration for CLI coding agents with HMAC-chained audit logs and signed agent cards. **When to choose:** Choose Bernstein when you need audit trails for AI-driven coding workflows, especially in regulated environments requiring compliance with standards like DORA, NIS2, or EU AI Act Article 12. **When NOT to choose:** Don't choose Bernstein if you need a SaaS solution with vendor support, if your use case doesn't require audit capabilities, or if you're building non-coding LLM workflows. ### [octocode](https://mcpcatalogs.com/en/server/muvon-octocode) by Muvon · score 52 · ★376 MCP server that transforms codebases into searchable knowledge graphs with semantic search and structural understanding. **When to choose:** Choose Octocode when you need your AI assistant to deeply understand codebase structure, dependencies, and relationships rather than just searching text. **When NOT to choose:** Don't choose Octocode if you need minimal external dependencies or can't set up API keys for embedding providers. ### [agent](https://mcpcatalogs.com/en/server/1mcp-app-agent) by 1mcp-app · score 52 · ★438 1MCP is a unified MCP runtime that aggregates multiple MCP servers into one, offering CLI mode for progressive tool discovery. **When to choose:** Choose 1MCP when you need to manage multiple MCP servers across different clients while maintaining a single configuration and reducing each agent's initial tool surface. **When NOT to choose:** Avoid 1MCP if you need minimal overhead or are building a simple MCP setup with only one or two servers, as it adds a layer of complexity that may not be necessary. ### [apktool-mcp-server](https://mcpcatalogs.com/en/server/zinja-coder-apktool-mcp-server) by zinja-coder · score 52 · ★437 A production-ready MCP server that enables Android APK analysis and reverse engineering with AI assistants. **When to choose:** Choose this MCP server when you need to perform reverse engineering tasks on Android APKs using AI assistance, especially for vulnerability analysis and code modification workflows. **When NOT to choose:** Avoid if you need Java-level decompilation (consider jadx-mcp-server instead) or if you're working with non-Android applications. ### [director](https://mcpcatalogs.com/en/server/fdmtl-director) by fdmtl · score 51 · ★478 Director is a local-first platform for creating and managing MCP playbooks that enhance AI agents with new skills and capabilities. **When to choose:** Choose Director when you need to manage multiple MCP servers with different configurations, want to bundle tools and prompts into portable playbooks, or need centralized authentication and logging for MCP services. **When NOT to choose:** Avoid Director if you need a lightweight, minimal MCP integration without additional management overhead, or if you prefer working directly with individual MCP servers rather than a management layer. ### [storybloq](https://mcpcatalogs.com/en/server/storybloq-storybloq) by Storybloq · score 51 · ★520 A comprehensive MCP server that provides cross-session context persistence for AI coding through tickets, issues, and handovers. **When to choose:** Choose Storybloq when you need structured, persistent context for AI coding sessions, especially when working on complex projects with multiple tickets and handoffs. **When NOT to choose:** Don't choose Storybloq if you need write access to your AI client (it's read-only), or if you prefer a simpler workflow without structured ticket management. ### [tableau-mcp](https://mcpcatalogs.com/en/server/tableau-tableau-mcp) by tableau · score 51 · ★270 Tableau's official MCP Server provides tools, resources and prompts for AI applications to integrate with Tableau data. **When to choose:** Choose this when you need AI agents to analyze, query and visualize Tableau data without leaving your AI interface. **When NOT to choose:** Avoid if you need write access to Tableau data or if you're not already using Tableau as your primary BI platform. ### [drift](https://mcpcatalogs.com/en/server/dadbodgeoff-drift) by dadbodgeoff · score 51 · ★781 Codebase intelligence tool that detects patterns and remembers decisions across sessions for AI assistants. **When to choose:** Choose Drift when you need AI assistants to understand your project's specific coding patterns and maintain context across development sessions. **When NOT to choose:** Avoid Drift if you need write access capabilities or require extensive documentation, as it appears focused on read-only code analysis with limited public documentation. ### [narsil-mcp](https://mcpcatalogs.com/en/server/postrv-narsil-mcp) by postrv · score 51 · ★151 Rust MCP server providing 90+ code intelligence tools across 32 languages with neural search and security analysis. **When to choose:** Choose narsil-mcp when you need comprehensive code analysis across multiple languages with security scanning and neural search capabilities, especially if you're working with large codebases and want offline processing. **When NOT to choose:** Avoid narsil-mcp if you only need basic LSP functionality without the heavy code analysis tools, or if you need real-time collaboration features as it processes everything locally. ### [postman-mcp-server](https://mcpcatalogs.com/en/server/postmanlabs-postman-mcp-server) by postmanlabs · score 51 · ★240 Postman MCP Server connects AI tools to Postman APIs for testing, collections management, and code generation. **When to choose:** Choose this server if you're already using Postman for API development and want to enable AI agents to interact with your API collections, environments, and documentation directly. **When NOT to choose:** Avoid if you're not using Postman, need to test private APIs with the remote server, or require custom authentication methods beyond OAuth or Postman API keys. ### [jadx-mcp-server](https://mcpcatalogs.com/en/server/zinja-coder-jadx-mcp-server) by zinja-coder · score 51 · ★487 An MCP server for JADX-AI Plugin that enables reverse engineering of Android APKs through LLM tool calls. **When to choose:** Choose this MCP server when you need deep, AI-powered analysis of Android APKs with JADX decompiler integration. **When NOT to choose:** Don't choose this if you need to analyze iOS apps, if you prefer GUI-only tools, or if you're not comfortable with CLI-based MCP setup. ### [flux-operator](https://mcpcatalogs.com/en/server/controlplaneio-fluxcd-flux-operator) by controlplaneio-fluxcd · score 51 · ★628 Flux MCP server bridges AI assistants with Kubernetes clusters for GitOps operations via natural language. **When to choose:** Choose this MCP server if you're already using Flux CD for GitOps operations and want to leverage AI assistants for analyzing deployments, troubleshooting issues, and performing operations through natural language. **When NOT to choose:** Avoid this if you're not using Flux CD in your infrastructure, or if you require GitOps tooling for platforms other than GitHub, GitLab, Gitea, and Azure DevOps without the Flux Operator. ### [unity-mcp-server](https://mcpcatalogs.com/en/server/anklebreaker-studio-unity-mcp-server) by AnkleBreaker-Studio · score 51 · ★182 Comprehensive MCP server with 288 tools for AI-assisted Unity development through Editor and Hub. **When to choose:** Unity developers who want to leverage AI assistants for complex game development tasks like scene creation, asset management, and automated testing. **When NOT to choose:** Teams not using Unity or those requiring extensive write access to project files beyond what Unity's editor API provides. ### [mcpproxy-go](https://mcpcatalogs.com/en/server/smart-mcp-proxy-mcpproxy-go) by smart-mcp-proxy · score 51 · ★218 MCPProxy is a smart proxy server that federates multiple MCP servers with security, token savings, and tool discovery. **When to choose:** Choose MCPProxy when you need to scale AI agents beyond platform limits while maximizing token efficiency and requiring strong security protections against malicious MCP servers. **When NOT to choose:** Don't choose MCPProxy if you only need a simple MCP server connection without the overhead of a proxy, or if you're working in a strictly air-gapped environment without internet access. ### [lumen](https://mcpcatalogs.com/en/server/ory-lumen) by ory · score 51 · ★190 Local semantic code search engine that reduces AI agent token usage by up to 66% through efficient code indexing. **When to choose:** Choose Lumen when working with AI coding assistants and need to significantly reduce token usage and costs while maintaining code quality, especially for TypeScript/JavaScript codebases. **When NOT to choose:** Avoid Lumen if you need to work with closed-source embedding models or require cloud-based vector search capabilities. ### [Charles-mcp](https://mcpcatalogs.com/en/server/heizaheiza-charles-mcp) by heizaheiza · score 51 · ★233 A production-ready MCP server for integrating Charles Proxy with AI agents, enabling real-time traffic capture and analysis. **When to choose:** Choose this when you need AI agents to analyze HTTP traffic patterns, reverse engineer APIs, or perform security assessments with a structured approach that minimizes token usage while allowing deep inspection. **When NOT to choose:** Don't choose this if you need to analyze non-HTTP protocols, if Charles Proxy is not already part of your workflow, or if you require a simpler traffic inspection tool without reverse engineering capabilities. ### [guild](https://mcpcatalogs.com/en/server/mathomhaus-guild) by mathomhaus · score 51 · ★190 Go-based MCP server providing shared context and task coordination for AI coding agents with hybrid search. **When to choose:** Choose Guild when working with multiple AI coding agents on the same project who need to maintain context and avoid redundant work across different MCP clients. **When NOT to choose:** Don't choose Guild if you need cloud-based collaboration features, or if you prefer solutions that work across multiple machines rather than being limited to a single local machine. ### [FantasyPremierLeague](https://mcpcatalogs.com/en/server/joreilly-fantasypremierleague) by joreilly · score 51 · ★593 Kotlin Multiplatform Fantasy Premier League app with MCP server exposing player/fixture data. **When to choose:** Choose this if you need Fantasy Premier League data accessible through MCP and are already using Kotlin in your tech stack. **When NOT to choose:** Don't choose this if you need comprehensive league management features beyond player data access. ### [mcp-sequentialthinking-tools](https://mcpcatalogs.com/en/server/spences10-mcp-sequentialthinking-tools) by spences10 · score 51 · ★583 An MCP server for recording sequential reasoning steps with history, branching, and tool plan validation. **When to choose:** Choose this server when your AI agent needs to explicitly break down complex problems, maintain reasoning history, or validate tool recommendations during multi-step processes. **When NOT to choose:** Don't use for trivial requests as it adds overhead, or if you need automatic tool discovery rather than validation of pre-specified tools. ### [claude-code-skills](https://mcpcatalogs.com/en/server/levnikolaevich-claude-code-skills) by levnikolaevich · score 51 · ★465 A comprehensive plugin suite with bundled MCP servers for Claude Code, including hex-line, hex-graph, hex-ssh, and hex-research for enhanced code intelligence. **When to choose:** Choose these MCP servers when you need hash-verified file editing, code intelligence graphs, or remote SSH operations with reduced token usage in development workflows. **When NOT to choose:** Avoid if you need simple file operations without verification overhead, or if you're looking for a single MCP server rather than a comprehensive toolkit. ### [vibe](https://mcpcatalogs.com/en/server/mondaycom-vibe) by mondaycom · score 51 · ★645 MCP server for monday.com's Vibe Design System that helps discover component APIs and usage examples. **When to choose:** Teams building applications with the Vibe Design System who want AI assistance for component selection and usage. **When NOT to choose:** Teams not using the Vibe Design System, or those seeking a more general-purpose UI component assistance tool. ### [mcp-odoo](https://mcpcatalogs.com/en/server/tuanle96-mcp-odoo) by tuanle96 · score 51 · ★316 MCP Server for Odoo that provides 24 tools, prompts, and safe write workflows without Odoo-side setup. **When to choose:** Choose this MCP server if you need AI access to an existing Odoo instance without installing additional modules or requiring administrator permissions. **When NOT to choose:** Don't choose this if you need direct write access to Odoo without approval workflow, or if you're using an Odoo version older than 16. ### [UnrealClaude](https://mcpcatalogs.com/en/server/natfii-unrealclaude) by Natfii · score 51 · ★613 Claude Code CLI integration for Unreal Engine 5.7 with MCP server exposing 20+ tools for UE development. **When to choose:** Choose UnrealClaude if you're working with Unreal Engine 5.7 and want AI-powered coding assistance with direct integration into the editor. **When NOT to choose:** Don't choose this if you're not using Unreal Engine 5.7, as it's specifically tailored for this version and may not work with others. ### [forgemax](https://mcpcatalogs.com/en/server/postrv-forgemax) by postrv · score 51 · ★148 Forgermax is an innovative MCP gateway that collapses multiple servers into just 2 tools using sandboxed JavaScript execution. **When to choose:** Choose Forgemax when working with many MCP servers and experiencing context window limitations or high round-trip latency. **When NOT to choose:** Avoid Forgemax if you need direct tool schemas for model understanding or if you're working with a small, fixed set of MCP tools. ### [lunar](https://mcpcatalogs.com/en/server/thelunarcompany-lunar) by TheLunarCompany · score 51 · ★444 Lunar.dev is an API gateway for governance and security of AI agent workloads with MCP aggregation capabilities. **When to choose:** Choose Lunar.dev when you need to govern and secure API access across multiple AI agents and applications, particularly when you require centralized monitoring and cost optimization of third-party API usage. **When NOT to choose:** Don't choose Lunar.dev if you're looking for a simple, lightweight MCP server or if you need an open-source solution for production use without paid tier requirements. ### [mcp-local-rag](https://mcpcatalogs.com/en/server/shinpr-mcp-local-rag) by shinpr · score 51 · ★260 Local-first RAG server providing semantic + keyword search for code and technical docs via MCP or CLI. **When to choose:** Choose this when you need private, local RAG search for technical documentation with exact term matching capabilities. **When NOT to choose:** Don't choose this if you need cloud-hosted search, large-scale enterprise deployment, or integration with existing vector databases. ### [Android-MCP](https://mcpcatalogs.com/en/server/cursortouch-android-mcp) by CursorTouch · score 51 · ★602 Android-MCP enables AI agents to interact with Android devices via ADB for UI automation, testing and control. **When to choose:** Choose Android-MCP when you need AI to interact with Android devices for testing, automation, or development workflows and prefer not to rely on computer vision. **When NOT to choose:** Avoid if you need to support iOS devices, or if you require lower latency than the 2-4s action-to-action timing that Android-MCP provides. ### [dexto](https://mcpcatalogs.com/en/server/truffle-ai-dexto) by truffle-ai · score 51 · ★620 A TypeScript-based agent harness that includes MCP server capabilities for building and orchestrating AI applications. **When to choose:** Choose Dexto when you need a comprehensive agent framework for building autonomous coding agents and multi-agent systems with rich tool integration and persistent memory. **When NOT to choose:** Consider alternatives if you need a more focused MCP server for specific tools rather than a full agent framework, or if you prefer a different license as Dexto uses Elastic 2.0 which may have restrictions for some organizations. ### [jdocmunch-mcp](https://mcpcatalogs.com/en/server/jgravelle-jdocmunch-mcp) by jgravelle · score 51 · ★171 MCP server for efficient documentation exploration via structured section indexing, reducing token consumption by 95% compared to traditional approaches. **When to choose:** When working with large documentation sets and need efficient, structured retrieval without flooding context windows. **When NOT to choose:** For teams with limited budgets as commercial use requires paid licensing, or if you need basic file access without section indexing. ### [mcp-shrimp-task-manager](https://mcpcatalogs.com/en/server/cjo4m06-mcp-shrimp-task-manager) by cjo4m06 · score 51 · ★2,100 MCP server for intelligent task management that breaks down complex projects into structured dev tasks with dependency tracking. **When to choose:** Choose Shrimp when working with AI agents on complex software development projects that require task breakdown, dependency management, and persistent context across sessions. **When NOT to choose:** Avoid if you need simple task management without AI integration, or if you prefer a lightweight solution without the overhead of a dedicated MCP server. ### [mcp-server-code-execution-mode](https://mcpcatalogs.com/en/server/elusznik-mcp-server-code-execution-mode) by elusznik · score 51 · ★331 An MCP server that executes Python code in isolated containers with MCP server proxying to reduce context bloat. **When to choose:** Choose this when managing multiple MCP servers and facing context window limitations, or when you need secure Python execution with scientific libraries and persistent state. **When NOT to choose:** Don't choose this if you need JavaScript/TypeScript execution, or if you require direct tool access without code abstraction layer. ### [lsbot](https://mcpcatalogs.com/en/server/ruilisi-lsbot) by ruilisi · score 50 · ★409 A lean, secure bot with MCP protocol integration and end-to-end encryption for AI applications. **When to choose:** Choose lsbot when you need a secure AI bot for enterprise messaging platforms with end-to-end encryption and comprehensive system access via MCP tools. **When NOT to choose:** Avoid lsbot if you need extensive third-party integrations beyond its supported platforms or require cloud-based conversation history and data storage. ### [EDT-MCP](https://mcpcatalogs.com/en/server/ditrixnew-edt-mcp) by DitriXNew · score 50 · ★156 MCP server for 1C:EDT that enables AI assistants to interact with development workspace through 56 specialized tools. **When to choose:** Choose EDT-MCP if you're working with 1C:EDT and want AI assistants to deeply understand your project structure, validate code, and perform metadata operations without context switching. **When NOT to choose:** Avoid this if you're not using 1C:EDT as it's specifically designed for this platform and requires the IDE to be running. ### [Mantic.sh](https://mcpcatalogs.com/en/server/marcoaapfortes-mantic-sh) by marcoaapfortes · score 50 · ★547 A context-aware code search engine for AI agents with semantic understanding of code structure. **When to choose:** Choose Mantic when you need context-aware code search for AI agents, require understanding code structure beyond simple text matching, and prioritize result quality over raw search speed. **When NOT to choose:** Avoid Mantic for very large repos (100K+ files) where maximum speed is critical, or for simple exact string searches where faster tools like ripgrep are sufficient. ### [winremote-mcp](https://mcpcatalogs.com/en/server/dddabtc-winremote-mcp) by dddabtc · score 50 · ★118 Windows Remote MCP Server with 40+ tools for desktop automation, process management, and file operations via FastMCP. **When to choose:** Choose this when you need comprehensive Windows desktop automation for AI agents, especially when integrating with Claude Desktop or OpenClaw for remote control of Windows machines. **When NOT to choose:** Avoid this if you need cross-platform support beyond Windows, require write access to Windows registries without tier restrictions, or need more than basic authentication for high-security environments. ### [mcp-pointer](https://mcpcatalogs.com/en/server/etsd-tech-mcp-pointer) by etsd-tech · score 50 · ★574 MCP Pointer enables AI assistants to see and analyze specific DOM elements through browser selection via MCP protocol. **When to choose:** When working on frontend development and need AI assistance to analyze specific DOM elements. **When NOT to choose:** If you're not using Chrome-based browsers or don't need AI assistance with web development elements. ### [mcp-server-tree-sitter](https://mcpcatalogs.com/en/server/wrale-mcp-server-tree-sitter) by wrale · score 50 · ★303 MCP server providing code analysis capabilities using Tree-sitter for multiple programming languages. **When to choose:** Choose this server when you need deep code analysis capabilities with AST parsing and want to maintain context management for large codebases across multiple programming languages. **When NOT to choose:** Don't choose this if you need write access to codebases (this is read-only only) or if you're working with very specialized domain-specific languages not supported by Tree-sitter. ### [generator](https://mcpcatalogs.com/en/server/context-hub-generator) by context-hub · score 50 · ★321 CTX is an MCP server providing AI access to read, write, and analyze code across projects with custom tools. **When to choose:** Choose CTX when working with PHP projects or when you need AI assistants with direct filesystem access and the ability to execute custom commands across multiple projects. **When NOT to choose:** Don't choose CTX if you need AI to work with languages other than PHP, or if you require a more comprehensive IDE integration beyond MCP connections. ### [claude-historian-mcp](https://mcpcatalogs.com/en/server/vvkmnn-claude-historian-mcp) by Vvkmnn · score 50 · ★179 An MCP server that allows Claude Code to search and retrieve conversation history with sophisticated search capabilities and zero dependencies. **When to choose:** Choose Claude Historian if you're working with Claude Code and need to efficiently search through past conversations, find solutions to recurring problems, or understand how specific files have been modified over time. **When NOT to choose:** Don't choose this if you need to work outside of Claude Code, as this MCP server is specifically designed for and only works with the Claude Code environment. ### [iterm-mcp](https://mcpcatalogs.com/en/server/ferrislucas-iterm-mcp) by ferrislucas · score 50 · ★556 MCP server that gives models control over your iTerm terminal with efficient token usage and full REPL support. **When to choose:** Choose iterm-mcp when you need AI assistance directly in your terminal environment, particularly for REPL interactions, CLI debugging, or automating terminal workflows. **When NOT to choose:** Avoid this if you need security restrictions on commands, work on non-macOS systems with iTerm, or require terminal interaction outside of the MCP protocol ecosystem. ### [one-mcp](https://mcpcatalogs.com/en/server/burugo-one-mcp) by burugo · score 50 · ★390 One MCP is a centralized proxy platform for managing multiple MCP services with a web interface, analytics, and service grouping capabilities. **When to choose:** Choose One MCP when you need to manage multiple MCP services from different sources through a unified interface, especially when you want to group services and export them as Anthropic Skills. **When NOT to choose:** Avoid One MCP if you need a lightweight, single-purpose MCP server with minimal overhead, as this is a comprehensive management platform with more complexity than simple MCP implementations. ### [iam-policy-autopilot](https://mcpcatalogs.com/en/server/awslabs-iam-policy-autopilot) by awslabs · score 50 · ★359 AWS IAM Policy Autopilot helps AI coding assistants create baseline IAM policies through code analysis. **When to choose:** Choose this when developing AWS applications with multiple languages and want to automate IAM policy generation with AI coding assistants. **When NOT to choose:** Don't choose this if you need resource-based policies (like S3 bucket policies) or are working with non-AWS environments. ### [perfetto-mcp](https://mcpcatalogs.com/en/server/antarikshc-perfetto-mcp) by antarikshc · score 50 · ★178 An MCP server that transforms natural language prompts into Perfetto trace analyses for Android performance debugging. **When to choose:** Choose this server when you need to analyze Android performance traces with natural language queries and prefer not to write complex Perfetto SQL manually. **When NOT to choose:** Don't choose this if you're working with non-Android performance traces or need more general-purpose trace analysis capabilities beyond Perfetto. ### [robloxstudio-mcp](https://mcpcatalogs.com/en/server/boshyxd-robloxstudio-mcp) by boshyxd · score 50 · ★431 MCP server connecting AI assistants to Roblox Studio for game development. **When to choose:** Choose this if you're developing games in Roblox Studio and want AI assistance to automate routine tasks, analyze code, or perform bulk operations directly within your development environment. **When NOT to choose:** Don't choose this if you need AI capabilities for game engines other than Roblox, or if you require remote access to your game development environment without a local Roblox Studio installation. ### [token-optimizer-mcp](https://mcpcatalogs.com/en/server/ooples-token-optimizer-mcp) by ooples · score 50 · ★369 MCP server for intelligent token optimization in Claude Code with caching and compression features. **When to choose:** Choose this server if you're using Claude Code regularly and need to minimize token costs while maintaining functionality. **When NOT to choose:** Avoid if you need fine-grained control over token optimization or are not experiencing significant token costs with Claude Code. ### [unity-mcp-plugin](https://mcpcatalogs.com/en/server/anklebreaker-studio-unity-mcp-plugin) by AnkleBreaker-Studio · score 50 · ★97 Unity MCP Plugin bridges Unity Editor with Claude, Cursor & other AI assistants via 288 tools for game development. **When to choose:** Choose this MCP server if you're working with Unity and want comprehensive AI-assisted development capabilities, particularly if you need to automate complex tasks or have AI build complete game systems from scratch. **When NOT to choose:** Don't choose this if you're not using Unity, need a different game engine, or require authentication/remote access since the server only binds to localhost. ### [hookdeck-cli](https://mcpcatalogs.com/en/server/hookdeck-hookdeck-cli) by hookdeck · score 50 · ★357 Hookdeck CLI provides a comprehensive MCP server for managing webhooks and event gateway resources, offering free development tools with ai agent integration. **When to choose:** Teams already using or planning to use Hookdeck for webhook management who want AI agent integration with their event infrastructure. **When NOT to choose:** You need a pure self-hosted solution without vendor dependencies, or you're looking for more generic webhook forwarding without MCP integration. ### [claude-mermaid](https://mcpcatalogs.com/en/server/veelenga-claude-mermaid) by veelenga · score 50 · ★146 MCP server for rendering Mermaid diagrams with live reload functionality and export options. **When to choose:** Choose this when you need to create, edit, and iterate on Mermaid diagrams with live preview functionality, especially if you're already using Claude Code. **When NOT to choose:** Don't choose this if you need diagram features not supported by Mermaid, or if you're working in an environment where you can't install command-line tools. ### [kicad-mcp-pro](https://mcpcatalogs.com/en/server/oaslananka-kicad-mcp-pro) by oaslananka · score 50 · ★126 MCP server for KiCad PCB and schematic automation with project setup, validation gates, and manufacturing exports. **When to choose:** Choose this MCP server if you're working with KiCad projects and need AI assistance for PCB design automation, validation, and manufacturing workflows. **When NOT to choose:** Don't choose this if you're not using KiCad or need more generalized CAD tool support beyond electronic design. ### [context-sync](https://mcpcatalogs.com/en/server/intina47-context-sync) by Intina47 · score 50 · ★124 Local persistent memory store for LLM applications providing project continuity through MCP tools. **When to choose:** Choose Context Sync when working with multiple coding sessions on the same project and need persistent memory for project context, decisions, and constraints without relying on the LLM's short-term memory. **When NOT to choose:** Avoid if you need write access to Notion or require bidirectional data synchronization between your project and external systems. ### [rails-ai-context](https://mcpcatalogs.com/en/server/crisnahine-rails-ai-context) by crisnahine · score 50 · ★139 38 MCP tools providing AI agents with direct access to Rails app schema, models, routes & conventions. **When to choose:** Choose this when working with Rails applications and want to reduce AI's guessing game by giving it direct, verified access to your schema, models, and conventions. **When NOT to choose:** Don't choose this if you're not using Rails, or if you need tools that can write or modify code (this server is strictly read-only). ### [Gopeak-godot-mcp](https://mcpcatalogs.com/en/server/had0yun-gopeak-godot-mcp) by HaD0Yun · score 50 · ★182 GoPeak is a comprehensive MCP server for Godot 4 with 95+ tools for game development workflows. **When to choose:** Choose GoPeak when working with Godot Engine and you need AI assistance with scene management, debugging, and game development workflows. **When NOT to choose:** Don't choose GoPeak if you're not using Godot Engine, or if you prefer a self-contained MCP server without external plugin dependencies. ### [UnityMCP](https://mcpcatalogs.com/en/server/isuzu-shiranui-unitymcp) by isuzu-shiranui · score 50 · ★131 Unity Editor integration with MCP enabling AI assistants to directly interact with Unity projects via HTTP and MCP tools. **When to choose:** Choose UnityMCP when you need to connect AI assistants directly with Unity development workflows, especially if you're already using Claude and need code execution capabilities. **When NOT to choose:** Don't choose UnityMCP if you need broader AI model support beyond Claude, require network exposure (it's loopback-only), or need stable versions across Unity version updates. ### [mcp-server-salesforce](https://mcpcatalogs.com/en/server/tsmztech-mcp-server-salesforce) by tsmztech · score 50 · ★156 MCP server that enables Claude to interact with Salesforce objects, records, metadata, and Apex code. **When to choose:** Choose this server if you need to interact with Salesforce data through AI and want a comprehensive set of tools for CRM object and record management. **When NOT to choose:** Avoid this if you're not using Salesforce, or if you need access to non-Salesforce data sources as this server is exclusively focused on Salesforce integration. ### [ckb](https://mcpcatalogs.com/en/server/nyxcore-systems-ckb) by nyxCore-Systems · score 50 · ★95 CKB is a comprehensive code intelligence platform providing MCP server with tools for code analysis, impact assessment, and architecture mapping. **When to choose:** Choose CKB when you need deep code intelligence for AI-assisted development, especially for Go codebases or when working with TypeScript/JavaScript/Python projects that require impact analysis and architecture mapping. **When NOT to choose:** Avoid CKB if you're working with unsupported languages (C#, PHP) or need non-Go incremental indexing, as the tool's effectiveness varies significantly by language tier and reindexing capabilities. ### [cross-code-organizer](https://mcpcatalogs.com/en/server/mcpware-cross-code-organizer) by mcpware · score 50 · ★318 Cross-platform configuration dashboard for AI coding tools including Claude Code and Codex CLI with MCP server management. **When to choose:** Choose Cross-Code Organizer if you use multiple AI coding tools (Claude Code, Codex CLI) and need a unified way to manage configurations, detect security issues in MCP servers, and optimize context usage. **When NOT to choose:** Don't choose this if you only use a single AI coding tool without MCP servers, or if you need deep integration with specific IDEs beyond what this dashboard provides. ### [claude-debugs-for-you](https://mcpcatalogs.com/en/server/jasonjmcghee-claude-debugs-for-you) by jasonjmcghee · score 50 · ★510 An MCP server enabling Claude or any LLM to interactively debug code in any language through VS Code. **When to choose:** Choose this when you want an AI-powered debugging assistant that works across multiple programming languages through VS Code and integrates with your preferred MCP client. **When NOT to choose:** Avoid this if you need debugging capabilities outside of VS Code or prefer traditional debugging methods without AI assistance. ### [memtrace-public](https://mcpcatalogs.com/en/server/syncable-dev-memtrace-public) by syncable-dev · score 50 · ★169 Memtrace provides structural memory for AI coding agents through MCP, indexing codebases as bi-temporal knowledge graphs with zero LLM calls. **When to choose:** Choose Memtrace when working with complex codebases where multiple AI coding agents need shared structural memory with temporal awareness and impact analysis capabilities. **When NOT to choose:** Don't choose Memtrace if you need open-source licensing (this is proprietary EULA), conversational memory capabilities beyond code, or if you're already deeply invested in alternative memory systems like Mem0 or Graphiti. ### [LeanKG](https://mcpcatalogs.com/en/server/freepeak-leankg) by FreePeak · score 50 · ★188 LeanKG is a local-first knowledge graph that indexes codebases and exposes them via MCP for AI-assisted development with significant token savings. **When to choose:** Choose LeanKG when you need a local, self-hosted knowledge graph to improve AI coding tools' accuracy with codebase context without cloud dependencies. **When NOT to choose:** Don't choose LeanKG if you need cloud-based collaboration features or require write access to the knowledge graph, as it's primarily a read-only system. ### [AI-company](https://mcpcatalogs.com/en/server/cronusl-1141-ai-company) by CronusL-1141 · score 50 · ★175 Autonomous multi-agent team OS for Claude Code with 108 MCP tools, 40+ agent templates, and real-time dashboard. **When to choose:** Choose this when you're already using Claude Code and want to create autonomous AI teams that can operate continuously without manual prompting, with structured workflows and comprehensive agent collaboration. **When NOT to choose:** Don't choose this if you need a solution that works with other AI platforms beyond Claude Code, or if you prefer simpler tools without the complexity of a full multi-agent operating system. ### [mcp-apache-spark-history-server](https://mcpcatalogs.com/en/server/kubeflow-mcp-apache-spark-history-server) by kubeflow · score 50 · ★170 MCP Server connecting AI agents to Apache Spark History Server for job analysis and performance monitoring. **When to choose:** Choose this MCP server if you're using Apache Spark and want AI agents to analyze cluster performance, debug applications, or compare job runs through natural language queries. **When NOT to choose:** Don't choose this if you're not using Apache Spark, need real-time streaming analytics, or require access to non-Hadoop ecosystem tools. ### [mcp-server-excel](https://mcpcatalogs.com/en/server/sbroenne-mcp-server-excel) by sbroenne · score 50 · ★157 MCP server that provides 23 tools and 214 operations for AI-powered Excel automation via COM API. **When to choose:** Choose this MCP server if you need AI-powered Excel automation on Windows with extensive functionality for data analysis, reporting, and workbook management. **When NOT to choose:** Avoid this server if you're using Linux/macOS, need server-side processing, or require high-volume batch operations without Excel installation. ### [touchdesigner-mcp](https://mcpcatalogs.com/en/server/8beeeaaat-touchdesigner-mcp) by 8beeeaaat · score 50 · ★324 An MCP server enabling AI agents to control TouchDesigner through node manipulation and Python script execution. **When to choose:** Choose this server if you're working with TouchDesigner projects and want to integrate AI for generative design, automated node manipulation, or creative programming workflows. **When NOT to choose:** Don't choose this if you need to work with other creative coding environments like Processing or p5.js, or if you require authentication for multi-user environments. ### [Axon.MCP.Server](https://mcpcatalogs.com/en/server/ali-kamali-axon-mcp-server) by ali-kamali · score 50 · ★165 MCP server that transforms codebases into intelligent knowledge bases for AI IDEs like Cursor and AntiGravity with semantic analysis and vector search. **When to choose:** Choose Axon when you need deep semantic understanding of your C# or Python codebase for AI-powered development in Cursor or other MCP-enabled IDEs. **When NOT to choose:** Not suitable if you need multi-language support beyond C# and Python, or if you prefer a lightweight solution without the overhead of its full microservice architecture. ### [chiasmus](https://mcpcatalogs.com/en/server/yogthos-chiasmus) by yogthos · score 50 · ★178 Chiasmus is an MCP server that provides LLMs with formal verification capabilities using Z3 and Prolog. **When to choose:** Choose Chiasmus when you need mathematically certain verification of code properties, security vulnerabilities, or system behavior beyond testing. **When NOT to choose:** Avoid if you need only basic code analysis or if your use case doesn't require formal mathematical verification capabilities. ### [claudeus-wp-mcp](https://mcpcatalogs.com/en/server/deus-h-claudeus-wp-mcp) by deus-h · score 50 · ★142 A comprehensive MCP server with 145 WordPress management tools for content, media, users, and site operations. **When to choose:** Choose this server when you need comprehensive WordPress site management through AI workflows, especially if you're already using Astra Pro theme or require full site editing capabilities. **When NOT to choose:** Avoid if you're not using WordPress or if you need a lightweight solution with fewer than 50 specialized tools. ### [glade-mcp-unity](https://mcpcatalogs.com/en/server/glade-tool-glade-mcp-unity) by Glade-tool · score 50 · ★132 A comprehensive Unity MCP server with 235+ tools, script search, and asset pipeline. **When to choose:** Choose GladeKit Unity MCP if you're working in Unity and want extensive game development tools with AI assistance directly in your workflow, especially for teams already using multiple AI clients. **When NOT to choose:** Don't choose this if you're not using Unity, or if you prefer a minimal toolset with fewer specialized game development features. ### [k8s-mcp-server](https://mcpcatalogs.com/en/server/reza-gholizade-k8s-mcp-server) by reza-gholizade · score 50 · ★156 A Go-based MCP server that provides comprehensive Kubernetes cluster management tools through MCP protocol. **When to choose:** Choose this when you need comprehensive Kubernetes management through AI interfaces and already have existing Kubernetes infrastructure to integrate with. **When NOT to choose:** Avoid if you need fine-grained RBAC control beyond what's provided, or if you're working with non-Kubernetes container orchestration platforms. ### [llm-context.py](https://mcpcatalogs.com/en/server/cyberchitta-llm-context-py) by cyberchitta · score 50 · ★300 Smart context management tool that allows sharing relevant project files with LLMs via MCP or clipboard with rule-based customization. **When to choose:** Choose this when you need intelligent, task-specific code context management with both human-friendly commands and MCP integration for AI agents. **When NOT to choose:** Avoid if you need read/write file access (this is read-only) or prefer simpler context sharing without rule-based configuration. ### [SharpToolsMCP](https://mcpcatalogs.com/en/server/kooshi-sharptoolsmcp) by kooshi · score 50 · ★197 A comprehensive MCP server for C# development with Roslyn-powered code analysis and modification capabilities. **When to choose:** Choose SharpTools when working with C# codebases and need AI to perform deep analysis, refactoring, or code modifications beyond simple text operations. **When NOT to choose:** Avoid if you're not working with C# code, or if you need a more general-purpose code analysis tool that supports multiple programming languages. ### [mcp_flutter](https://mcpcatalogs.com/en/server/arenukvern-mcp-flutter) by Arenukvern · score 50 · ★294 MCP server for Flutter AI agents to inspect, interact with, and control running Flutter applications. **When to choose:** Choose this when working with Flutter apps and need AI agents to interact with UI elements, take screenshots, or debug running applications through a closed feedback loop. **When NOT to choose:** Avoid if you need to interact with native iOS/Android code directly, or if your workflow requires more general-purpose app testing tools not specific to Flutter. ### [django-mcp-server](https://mcpcatalogs.com/en/server/gts360-django-mcp-server) by gts360 · score 50 · ★332 Django MCP Server enables AI agents to interact with Django applications through Model Context Protocol. **When to choose:** Choose this if you're working with Django applications and want to enable AI agents to interact with your models and APIs through MCP. **When NOT to choose:** Avoid if you need complex write operations or if you're not using Django, as this server is specifically designed for Django applications. ### [kodit](https://mcpcatalogs.com/en/server/helixml-kodit) by helixml · score 49 · ★119 Kodit is a code and document intelligence server that indexes Git repositories and provides search through MCP and REST APIs. **When to choose:** Choose Kodit when you need a self-hosted solution for indexing your own repositories and want to give AI assistants direct access to your codebase patterns without sending code to external services. **When NOT to choose:** Avoid Kodit if you need write access to repositories (this is read-only) or if you're looking for a cloud-hosted solution with minimal infrastructure management. ### [mcp-svelte-docs](https://mcpcatalogs.com/en/server/spences10-mcp-svelte-docs) by spences10 · score 49 · ★122 MCP server providing authoritative Svelte 5 documentation with comprehensive definitions and examples. **When to choose:** Developers working with Svelte 5 who need authoritative, TypeScript-based documentation in their AI environment without switching contexts. **When NOT to choose:** Teams using Svelte 4 who don't need migration guidance or those seeking documentation for frameworks other than Svelte. ### [sudocode](https://mcpcatalogs.com/en/server/sudocode-ai-sudocode) by sudocode-ai · score 49 · ★277 Lightweight agent orchestration dev tool with MCP server integration for collaborative AI development. **When to choose:** Choose sudocode when you need structured coordination of multiple AI coding agents and want context management integrated directly into your git workflow. **When NOT to choose:** Don't choose sudocode if you prefer simpler tools or need integration with specific issue trackers not yet supported, as it adds complexity to your development workflow. ### [metorial-platform](https://mcpcatalogs.com/en/server/metorial-metorial-platform) by metorial · score 49 · ★207 Metorial is an enterprise-grade MCP platform that connects AI models to thousands of APIs and tools with one-liner SDKs. **When to choose:** Choose Metorial when you need a managed platform to deploy and scale MCP servers across your organization. **When NOT to choose:** Don't choose Metorial if you're looking for a specific MCP server implementation or prefer self-hosting individual MCP servers. ### [Pare](https://mcpcatalogs.com/en/server/dave-london-pare) by Dave-London · score 49 · ★121 MCP servers that transform CLI tools into structured JSON outputs for AI agents, reducing token usage by up to 95%. **When to choose:** Choose Pare when working with AI agents that need reliable, structured output from development tools, especially when token efficiency is important or when working across different platforms. **When NOT to choose:** Don't choose Pare if you need raw terminal output with full formatting, or if your workflow depends on parsing specific terminal text patterns that would be lost in the JSON conversion. ### [open-computer-use](https://mcpcatalogs.com/en/server/wide-moat-open-computer-use) by Wide-Moat · score 49 · ★74 MCP server providing Docker workspaces with browser, terminal, code execution and document skills for any LLM. **When to choose:** Choose this when you need a self-hosted, production-ready computer environment for any LLM with strong isolation, multi-user support, and rich document creation capabilities. **When NOT to choose:** Avoid if you need simpler solutions, minimal resource usage, or are not comfortable with Docker-based architectures and complex setup processes. ### [android-source-explorer-mcp](https://mcpcatalogs.com/en/server/mrmike-android-source-explorer-mcp) by mrmike · score 49 · ★91 An MCP server providing direct access to Android source code for AI-assisted development. **When to choose:** Choose this when developing Android applications and you need precise, up-to-date information directly from the source code rather than potentially outdated documentation. **When NOT to choose:** Avoid if you're not working with Android development or if you need real-time access to the very latest AOSP changes before they're synced locally. ### [arbor](https://mcpcatalogs.com/en/server/anandb71-arbor) by Anandb71 · score 49 · ★117 A graph-native code intelligence tool with 10 MCP tools for analyzing code dependencies and impact. **When to choose:** Choose Arbor when you need deterministic, explainable code impact analysis across language boundaries, especially for large codebases where understanding blast radius before changes is critical. **When NOT to choose:** Don't choose Arbor if you only work with small, single-language projects where the overhead of maintaining a dependency graph outweighs the benefits. ### [agent-toolkit](https://mcpcatalogs.com/en/server/sanity-io-agent-toolkit) by sanity-io · score 49 · ★137 MCP server providing direct access to Sanity projects with content management tools and development best practices. **When to choose:** Choose this if you're already using Sanity CMS and want AI agents to directly interact with your content management system, schemas, and releases. **When NOT to choose:** Don't choose this if you're not using Sanity CMS, as it requires integration with their platform and provides no value for other CMS systems. ### [mcp-engine-public](https://mcpcatalogs.com/en/server/maxanatsko-mcp-engine-public) by maxanatsko · score 49 · ★237 MCP server for Power BI model interaction, enabling AI assistants to programmatically query and modify Power BI models. **When to choose:** Choose this if you work with Power BI models and want to use AI assistants to programmatically interact with them without leaving your preferred AI environment. **When NOT to choose:** Don't choose this if you need open-source solutions, as it uses a proprietary license, or if you're not using Power BI models. ### [remote-mcp-functions-python](https://mcpcatalogs.com/en/server/azure-samples-remote-mcp-functions-python) by Azure-Samples · score 49 · ★121 Azure Functions template for deploying MCP servers to the cloud with Python authentication options. **When to choose:** Choose this template if you're already using Azure services and want to deploy MCP servers in the cloud with enterprise authentication and scalability. **When NOT to choose:** Avoid this if you prefer self-hosted solutions, need offline capabilities, or are concerned about vendor lock-in to Azure services. ### [context-space](https://mcpcatalogs.com/en/server/context-space-context-space) by context-space · score 49 · ★809 Context Space provides unified MCP tools and verified integrations for AI agents with enterprise-grade security and a 5-minute setup. **When to choose:** Choose Context Space when you need unified access to multiple enterprise services through MCP for AI agents, especially when security and credential management are priorities. **When NOT to choose:** Avoid Context Space if you need a self-hosted solution without vendor lock-in, or if you prefer a more traditional MCP server approach rather than a unified API wrapper. ### [adloop](https://mcpcatalogs.com/en/server/klosk-adloop) by kLOsk · score 49 · ★155 AdLoop is an MCP server providing read/write access to Google Ads and GA4 with safety guardrails. **When to choose:** Choose AdLoop if you're managing Google Ads and GA4 campaigns and want AI assistance across both platforms with safety features to prevent accidental spending. **When NOT to choose:** Don't choose AdLoop if you need to manage advertising platforms other than Google Ads or require deep integration with non-Google analytics solutions. ### [Geargrafx](https://mcpcatalogs.com/en/server/drhelius-geargrafx) by drhelius · score 49 · ★213 PC Engine/TurboGrafx-16 emulator with embedded MCP server for AI-assisted debugging. **When to choose:** Choose Geargrafx MCP if you're a retro game developer, reverse engineer, or ROM hacker who wants to leverage AI assistants for debugging game code and analyzing game behavior in real-time. **When NOT to choose:** Don't choose this for modern game development, as it's specifically designed for retro systems. Also consider alternatives if you need a more generalized debugging tool or if you're not interested in TurboGrafx-16/PC Engine systems. ### [frontmcp](https://mcpcatalogs.com/en/server/agentfront-frontmcp) by agentfront · score 49 · ★143 TypeScript-first framework for building MCP servers with decorators, DI, and Streamable HTTP transport. **When to choose:** Choose FrontMCP when you want a TypeScript-based framework with decorators and dependency injection for building MCP servers with robust authentication and session management. **When NOT to choose:** Avoid FrontMCP if you need support for Node.js versions below 24 or prefer a simpler framework without the complexity of dependency injection. ### [freecad-addon-robust-mcp-server](https://mcpcatalogs.com/en/server/spkane-freecad-addon-robust-mcp-server) by spkane · score 49 · ★84 A comprehensive MCP server connecting AI assistants to FreeCAD for 3D modeling with 150+ tools. **When to choose:** Choose this when working with FreeCAD and want to leverage AI assistants for parametric modeling, automation, and debugging within your CAD workflow. **When NOT to choose:** Avoid if you're not using FreeCAD, as this is specifically designed for the FreeCAD ecosystem with no support for other CAD applications. ### [skillz](https://mcpcatalogs.com/en/server/intellectronica-skillz) by intellectronica · score 49 · ★394 MCP server for loading Claude-style skills as tools in any MCP client. **When to choose:** Choose Skillz when you want to use Claude-style skills in non-Claude AI clients or need more flexible skill organization than Claude Code provides. **When NOT to choose:** Don't choose Skillz if you're only working within Claude's ecosystem where native skill support already exists, or if you require extensive authentication/authorization models as it's designed for local skill management. ### [airbroke](https://mcpcatalogs.com/en/server/icoretech-airbroke) by icoretech · score 49 · ★215 Airbroke is a PostgreSQL-based error tracking service with MCP API integration for LLM/agent connections. **When to choose:** Choose Airbroke if you're already using PostgreSQL and want AI-powered error analysis without vendor lock-in. **When NOT to choose:** Avoid Airbroke if you need advanced analytics features not available in a lightweight solution or if you prefer managed cloud services over self-hosting. ### [gtm-mcp-server](https://mcpcatalogs.com/en/server/paolobietolini-gtm-mcp-server) by paolobietolini · score 49 · ★86 Google Tag Manager MCP server that enables natural language control of GTM containers through AI assistants. **When to choose:** Choose this if you're already using Google Tag Manager and want to manage it through natural language with AI assistants like Claude or ChatGPT. **When NOT to choose:** Don't choose this if you need GTM operations without an internet connection, as it requires connectivity to Google APIs and the MCP server. ### [nitrostack](https://mcpcatalogs.com/en/server/nitrocloudofficial-nitrostack) by nitrocloudofficial · score 49 · ★119 NitroStack is an enterprise-grade TypeScript framework for building production-ready MCP servers with decorators, dependency injection, and UI widgets. **When to choose:** Choose NitroStack when you need a full-stack TypeScript framework with enterprise features like authentication, dependency injection, and UI widgets for building production MCP servers. **When NOT to choose:** Avoid NitroStack if you're building a simple MCP server without need for advanced features, or if you prefer a more minimal approach without the framework's opinionated structure. ### [VectorCode](https://mcpcatalogs.com/en/server/davidyz-vectorcode) by Davidyz · score 49 · ★859 VectorCode is a code repository indexing tool that enhances LLM understanding of codebases through MCP integration. **When to choose:** Choose VectorCode when working with proprietary or cutting-edge codebases where LLMs typically lack understanding. **When NOT to choose:** Avoid if you need write access to repositories or require multi-language support beyond Python-based code analysis. ### [ue-mcp](https://mcpcatalogs.com/en/server/db-lyon-ue-mcp) by db-lyon · score 49 · ★96 Complete Unreal Engine development toolkit with 450+ MCP tools for AI-powered game development workflows. **When to choose:** Choose UE-MCP if you're working with Unreal Engine development and want to integrate AI assistants directly into your game creation workflow for automation and enhanced productivity. **When NOT to choose:** Don't choose UE-MCP if you need open-source licensing (it uses BUSL-1.1 with commercial licensing requirements) or if you're not using Unreal Engine in your development workflow. ### [agentic-tools](https://mcpcatalogs.com/en/server/meta-quest-agentic-tools) by meta-quest · score 49 · ★59 MCP server with 40+ tools for Meta Quest device management, app control, and VR development. **When to choose:** Choose this MCP server when developing or debugging Meta Quest VR applications and you want direct device control and performance analysis through your AI coding assistant. **When NOT to choose:** Don't choose this if you're not working with Meta Quest hardware or if you need MCP support for platforms outside the Quest ecosystem. ### [BifrostMCP](https://mcpcatalogs.com/en/server/biegehydra-bifrostmcp) by biegehydra · score 49 · ★212 VSCode extension exposing language server features to AI tools via MCP protocol. **When to choose:** Choose Bifrost when you need AI assistants to have deep code understanding and navigation capabilities within your VS Code environment. **When NOT to choose:** Avoid Bifrost if you need to work with languages that don't have VS Code language server support, or if you prefer a standalone IDE-agnostic solution. ### [augments-mcp-server](https://mcpcatalogs.com/en/server/augmnt-augments-mcp-server) by augmnt · score 49 · ★122 Comprehensive MCP server providing real-time documentation access for any npm package with intelligent search and context-aware formatting. **When to choose:** Choose augments when you need comprehensive documentation access across the entire npm ecosystem with intelligent search and context-aware formatting, especially when working with unfamiliar packages or migrating between versions. **When NOT to choose:** Don't choose augments if you only need documentation for a curated set of packages or prefer offline access, as it relies on real-time fetching from npm and GitHub. ### [ai-agent-team](https://mcpcatalogs.com/en/server/peterfei-ai-agent-team) by peterfei · score 49 · ★355 AI Agent Team is an MCP server providing a 24/7 AI development team with specialized roles for product management, frontend, backend, QA, DevOps, and technical leadership. **When to choose:** Choose this when you need persistent memory and context management across long development sessions, especially when working on complex projects that require switching between different development roles and maintaining conversation history. **When NOT to choose:** Avoid this if you need simple, stateless interactions or if you're working in a team environment where thread data isolation between projects is essential and the global sharing doesn't fit your workflow. ### [jebmcp](https://mcpcatalogs.com/en/server/flankerhqd-jebmcp) by flankerhqd · score 49 · ★235 MCP server enabling AI assistants to interact with JEB Pro reverse engineering tool for binary analysis. **When to choose:** Choose jebmcp when you're working with JEB Pro for Android reverse engineering and want to leverage AI assistants for automated analysis, code exploration, and batch renaming tasks. **When NOT to choose:** Avoid if you're not using JEB Pro, or if you need reverse engineering capabilities for platforms other than Android applications. ### [turbomcp](https://mcpcatalogs.com/en/server/epistates-turbomcp) by Epistates · score 49 · ★84 Production-ready Rust SDK for MCP with zero-boilerplate macros, modular transport architecture, and WASM support. **When to choose:** Choose TurboMCP for production MCP servers in Rust environments requiring multiple transport protocols, authentication, and comprehensive enterprise features. **When NOT to choose:** Avoid if you need a simple, lightweight solution or are working in a language other than Rust. ### [mcp-memory-keeper](https://mcpcatalogs.com/en/server/mkreyman-mcp-memory-keeper) by mkreyman · score 49 · ★122 MCP server for persistent context management in AI coding assistants that prevents context loss during session resets. **When to choose:** Choose this when working with Claude Code on long coding sessions or complex projects where preserving context between sessions is critical. **When NOT to choose:** Not suitable if you're using AI assistants other than Claude, as this is specifically designed for Claude's context management. ### [mcp-hosted](https://mcpcatalogs.com/en/server/forcedotcom-mcp-hosted) by forcedotcom · score 49 · ★109 Salesforce's hosted MCP servers enable AI assistants to securely connect to Salesforce assets using the Model Context Protocol. **When to choose:** Choose this if you need to connect AI assistants to Salesforce data and automation flows and want enterprise-grade authentication without building custom integrations. **When NOT to choose:** Avoid this if you're not using Salesforce, as it requires a Salesforce instance and only works with MCP-compatible AI clients. ### [xclaude-plugin](https://mcpcatalogs.com/en/server/conorluddy-xclaude-plugin) by conorluddy · score 49 · ★152 Modular iOS development automation with 8 workflow-specific MCP servers for Claude Code, enabling focused token usage and efficient workflows. **When to choose:** Choose xclaude-plugin for iOS development teams using Claude who want workflow-specific tools with minimal token usage and composable development loops. **When NOT to choose:** Not suitable for teams not using macOS/Linux, those not on Xcode 15+, or developers who prefer monolithic tools over modular architectures. ### [gram](https://mcpcatalogs.com/en/server/speakeasy-api-gram) by speakeasy-api · score 49 · ★233 Gram is a platform for creating, curating, and hosting MCP servers with TypeScript SDK and OpenAPI support. **When to choose:** Choose Gram when you need a managed platform for creating and hosting MCP servers at scale, especially if you're already working with TypeScript or OpenAPI specs. **When NOT to choose:** Avoid Gram if you prefer self-hosted solutions without vendor lock-in, or if you need simple MCP server implementation without the additional platform features. ### [iai-mcp](https://mcpcatalogs.com/en/server/codeabra-iai-mcp) by CodeAbra · score 49 · ★120 IAI-MCP is a high-performance memory system for AI coding assistants that implements MCP protocol to provide long-term memory. **When to choose:** Choose IAI-MCP when you need long-term memory for your AI coding assistant with verbatim recall, low latency, and complete privacy with local-only storage. **When NOT to choose:** Don't choose IAI-MCP if you're not on macOS, need cross-platform support, or require memory systems that work with non-MCP AI assistants. ### [pwno-mcp](https://mcpcatalogs.com/en/server/pwno-io-pwno-mcp) by pwno-io · score 49 · ★264 MCP server that exposes GDB+pwndbg debugging capabilities for binary research and exploit development. **When to choose:** Security researchers and reverse engineers who want LLM agents to assist with binary analysis and exploit development in an isolated environment. **When NOT to choose:** You shouldn't use this if you need commercial licensing without contacting the authors, or if you require architectures beyond what GDB+pwndbg supports. ### [mcp-server-azure-devops](https://mcpcatalogs.com/en/server/tiberriver256-mcp-server-azure-devops) by Tiberriver256 · score 49 · ★367 A well-implemented MCP server for Azure DevOps that provides comprehensive API integration through MCP protocol. **When to choose:** Choose this server if you're already using Azure DevOps and want to enable AI assistants to interact with your DevOps workflows through natural language commands. **When NOT to choose:** Don't choose this server if you're not using Azure DevOps or need to integrate with other DevOps platforms like GitHub, GitLab, or Jenkins. ### [deep-code-reasoning-mcp](https://mcpcatalogs.com/en/server/evalops-deep-code-reasoning-mcp) by evalops · score 49 · ★105 MCP server combining Claude Code with Google's Gemini AI for advanced code analysis and reasoning. **When to choose:** Choose this MCP server when your development workflow requires deep code analysis beyond single-file contexts, especially for distributed systems, performance regression hunting, or when you need to correlate failures across multiple services and logs. **When NOT to choose:** Avoid this server if you cannot use Google's Gemini API due to compliance or vendor lock-in concerns, or if you primarily work with small codebases that don't require the massive context window capabilities. ### [utcp-mcp](https://mcpcatalogs.com/en/server/universal-tool-calling-protocol-utcp-mcp) by universal-tool-calling-protocol · score 49 · ★198 A universal MCP bridge that connects AI agents to any native endpoint via UTCP with zero installation required. **When to choose:** Choose UTCP-MCP when you need to connect AI agents to diverse APIs and tools without writing custom MCP implementations, especially when working across multiple environments. **When NOT to choose:** Avoid if you need strict authentication and security controls, as the current implementation appears limited in enterprise-grade security features. ### [mcp-hub](https://mcpcatalogs.com/en/server/ravitemer-mcp-hub) by ravitemer · score 49 · ★487 A centralized hub for managing multiple MCP servers with unified endpoints and dynamic server monitoring. **When to choose:** Choose MCP Hub when you need to manage multiple MCP servers with a centralized interface and want to simplify client connections through a single endpoint. **When NOT to choose:** Avoid MCP Hub if you're running a simple setup with only one MCP server, as it adds unnecessary complexity for basic use cases. ### [godot-mcp](https://mcpcatalogs.com/en/server/tugcantopaloglu-godot-mcp) by tugcantopaloglu · score 48 · ★189 Godot MCP server with 149 tools for full engine control in AI-driven game development. **When to choose:** Choose Godot MCP when you need AI to perform complex Godot game development tasks, automate workflows, or create AI-powered game development experiences. **When NOT to choose:** Don't choose this if you're working with game engines other than Godot, or if you need a solution with GUI-based visual scripting capabilities. ### [bitbucket-mcp](https://mcpcatalogs.com/en/server/matanyemini-bitbucket-mcp) by MatanYemini · score 48 · ★136 Bitbucket MCP server enables AI assistants to interact with Bitbucket repositories and pull requests. **When to choose:** Choose this MCP server if your team uses Bitbucket for version control and you want AI assistants to directly interact with repositories, pull requests, and code review processes. **When NOT to choose:** Avoid this if you need write operations to Bitbucket (limited to PRs only) or if you use GitHub exclusively, as there's no equivalent GitHub integration from this author. ### [Dataverse-skills](https://mcpcatalogs.com/en/server/microsoft-dataverse-skills) by microsoft · score 48 · ★101 Microsoft Dataverse skills for AI coding agents that wrap Dataverse MCP server, CLI, SDK and PAC CLI. **When to choose:** Choose this if you're working with Microsoft Dataverse, Dynamics 365, or Power Platform and want AI agents to manage and query your data environments without requiring direct CLI or API expertise. **When NOT to choose:** Avoid this if you're not using Microsoft's Dataverse ecosystem, as it requires an existing Dataverse environment and doesn't support alternative data platforms. ### [unreal-api-mcp](https://mcpcatalogs.com/en/server/codeturion-unreal-api-mcp) by Codeturion · score 48 · ★75 MCP server providing instant, accurate Unreal Engine API documentation to prevent hallucinations and save tokens. **When to choose:** Developers working with Unreal Engine who need accurate API lookups to prevent hallucinations and reduce token usage. **When NOT to choose:** Third-party plugin or marketplace asset development (only built-in Engine and Plugin APIs are covered), or if you need write access to API documentation. ### [Pepper](https://mcpcatalogs.com/en/server/skwallace36-pepper) by skwallace36 · score 48 · ★88 Pepper is an iOS dynamic library MCP server that gives AI agents eyes and hands inside iOS Simulator apps. **When to choose:** Choose Pepper when you need AI-powered testing and debugging of iOS applications in Simulator without modifying source code or adding SDK dependencies. **When NOT to choose:** Don't choose Pepper if you need to test on physical devices, work with closed-source apps that prohibit instrumentation, or require cross-platform mobile automation. ### [trainingpeaks-mcp](https://mcpcatalogs.com/en/server/jamsusmaximus-trainingpeaks-mcp) by JamsusMaximus · score 48 · ★65 TrainingPeaks MCP server enables querying workouts, fitness data, and PRs via natural language. **When to choose:** Choose this server if you use TrainingPeaks for fitness tracking and want AI assistance for planning and analyzing your training without going through the API approval process. **When NOT to choose:** Don't choose this if you use a different fitness platform or need official API integration with advanced features not supported by cookie-based authentication. ### [template-mcp-server](https://mcpcatalogs.com/en/server/redhat-data-and-ai-template-mcp-server) by redhat-data-and-ai · score 48 · ★54 A production-ready Python template for building MCP servers with FastMCP, FastAPI, OAuth, and deployment examples. **When to choose:** Choose this template when starting a new MCP server project, especially if you need production-ready features like OAuth authentication, container deployment, and CI/CD integration. **When NOT to choose:** Don't choose this if you're building a simple proof-of-concept or if you need a different programming language (this template is Python-specific). ### [gdal-mcp](https://mcpcatalogs.com/en/server/jordangunn-gdal-mcp) by JordanGunn · score 48 · ★70 GDAL MCP provides AI agents with geospatial analysis capabilities through reflection middleware that justifies methodological choices. **When to choose:** Choose GDAL MCP when working with geospatial data requires documented methodology for reproducible science, or when AI agents need to understand spatial operations beyond just executing them. **When NOT to choose:** Don't choose GDAL MCP for simple geospatial tasks where methodology justification adds unnecessary overhead, or if you need write access capabilities (currently read-only). ### [mcp-remote-macos-use](https://mcpcatalogs.com/en/server/baryhuang-mcp-remote-macos-use) by baryhuang · score 48 · ★482 MCP server enabling AI agents to control remote macOS systems without API keys **When to choose:** Choose this when you need to control remote macOS systems through AI agents without API costs and prefer using native macOS applications. **When NOT to choose:** Don't choose this if you need authentication beyond Apple's protocol 30 or require a more secure 2048-bit RSA encryption. ### [mcp-toolkit](https://mcpcatalogs.com/en/server/nuxt-modules-mcp-toolkit) by nuxt-modules · score 48 · ★148 Nuxt module for creating MCP servers with automatic discovery of tools, resources, and prompts. **When to choose:** Choose this if you're already using Nuxt and want to integrate MCP functionality directly into your application with minimal setup. **When NOT to choose:** Avoid this if you're not using Nuxt or need a standalone MCP server solution that doesn't depend on the Nuxt framework. ### [mcp](https://mcpcatalogs.com/en/server/vuetifyjs-mcp) by vuetifyjs · score 48 · ★96 Vuetify MCP server provides Vuetify component documentation and API information to AI assistants via the Model Context Protocol. **When to choose:** Choose this MCP server if you're working with Vuetify UI components and want AI assistance directly in your development environment. **When NOT to choose:** Not suitable if you're not using Vuetify, or if you need write access to your component library (this server is read-only). ### [pitlane-mcp](https://mcpcatalogs.com/en/server/eresende-pitlane-mcp) by eresende · score 48 · ★73 Token-efficient code intelligence MCP server that indexes code once and lets agents retrieve only needed symbols. **When to choose:** Choose pitlane-mcp when working with large codebases where AI agents need efficient code navigation without context bloat, and when you need symbol-level retrieval rather than entire file analysis. **When NOT to choose:** Avoid pitlane-mcp if you need write access to the codebase (it's read-only) or if you're working with a small codebase where the indexing overhead isn't justified by the token savings. ### [netbox-mcp-server](https://mcpcatalogs.com/en/server/netboxlabs-netbox-mcp-server) by netboxlabs · score 48 · ★172 A well-maintained MCP server for read-only access to NetBox network infrastructure data via LLMs. **When to choose:** Choose this if you're already using NetBox for network infrastructure management and want LLMs to query your network data without writing custom integration code. **When NOT to choose:** Don't choose this if you need write access to NetBox, require advanced GraphQL features, or have complex authentication needs beyond API tokens. ### [dbt-doctor](https://mcpcatalogs.com/en/server/astoriel-dbt-doctor) by Astoriel · score 48 · ★134 AI-driven quality & governance MCP server for dbt projects, offering auditing, profiling, drift detection, and documentation generation. **When to choose:** Choose this when you want AI-powered dbt project health auditing, data profiling, and auto-documentation without leaving your AI assistant interface. **When NOT to choose:** Don't choose this if you need direct write access to your database or if you prefer traditional CLI-based dbt workflow without AI integration. ### [chrome-devtools-mcp](https://mcpcatalogs.com/en/server/benjaminr-chrome-devtools-mcp) by benjaminr · score 48 · ★297 Chrome DevTools MCP server enables web debugging through Claude by connecting to Chrome's debugging capabilities. **When to choose:** Choose this MCP server when you need to debug web applications through Claude, especially if you're already using Chrome DevTools and want to integrate browser debugging directly into your AI workflow. **When NOT to choose:** Don't choose this if you need to debug browsers other than Chrome, or if you require deep DOM manipulation capabilities beyond what Chrome DevTools provides. ### [mirroir-mcp](https://mcpcatalogs.com/en/server/jfarcand-mirroir-mcp) by jfarcand · score 48 · ★89 MCP server for controlling a real iPhone via macOS iPhone Mirroring with screen analysis and interaction capabilities. **When to choose:** Choose mirroir-mcp when you need AI agents to interact with iOS apps through a physical iPhone, especially for testing or automation tasks requiring real device interaction. **When NOT to choose:** Don't choose mirroir-mcp if you need Android support, don't have a Mac with iPhone Mirroring, or require non-UI automation like background data processing. ### [antigravity-link-extension](https://mcpcatalogs.com/en/server/cafetechne-antigravity-link-extension) by cafeTechne · score 48 · ★163 Mobile companion for Google's Antigravity IDE with 9 MCP tools and OpenAPI for automation. **When to choose:** Choose this if you're already using Google's Antigravity IDE and need mobile access or programmatic control of your AI sessions without leaving your local network. **When NOT to choose:** Don't choose this if you're not using Antigravity IDE, as it's specifically designed for this platform only. ### [mcp-dotnet-samples](https://mcpcatalogs.com/en/server/microsoft-mcp-dotnet-samples) by microsoft · score 48 · ★185 Microsoft's collection of MCP server samples in .NET for connecting AI models to various services. **When to choose:** Choose this if you're a .NET developer looking for examples of how to implement MCP servers for various services. **When NOT to choose:** Don't choose this if you're looking for a ready-to-use MCP server - these are sample implementations that require integration into your own applications. ### [dedalus-mcp-python](https://mcpcatalogs.com/en/server/dedalus-labs-dedalus-mcp-python) by dedalus-labs · score 48 · ★152 A minimal and spec-faithful Python framework for building MCP clients and servers with ergonomic decorators. **When to choose:** Choose Dedalus MCP when you need a spec-compliant, lightweight MCP implementation that integrates cleanly with your existing infrastructure without bundling unnecessary opinionated components. **When NOT to choose:** Don't choose Dedalus MCP if you need turnkey everything including auth providers, CLI scaffolding, and opinionated middleware - FastMCP would be more suitable for those needs. ### [tauri-plugin-mcp](https://mcpcatalogs.com/en/server/p3gleg-tauri-plugin-mcp) by P3GLEG · score 48 · ★96 Tauri plugin enabling AI agents to debug applications through screenshots, DOM access, and input simulation. **When to choose:** Choose this when you need AI-powered debugging and interaction specifically for Tauri applications, especially during development workflows. **When NOT to choose:** Don't choose this if you're not using Tauri, or if you need cross-desktop debugging capabilities beyond what Tauri provides. ### [cesium-mcp](https://mcpcatalogs.com/en/server/gaopengbin-cesium-mcp) by gaopengbin · score 48 · ★103 AI-powered CesiumJS 3D globe control with 49 tools for camera, entities, layers, animation & spatial analysis via MCP. **When to choose:** Choose Cesium MCP when you need AI-powered natural language control over CesiumJS 3D globes, especially for GIS applications, data visualization, or integrating AI assistants with geographic systems. **When NOT to choose:** Not suitable for projects requiring high-performance rendering of massive geographic datasets or those needing write access to CesiumJS internal state, as this server focuses on providing tools for AI interaction rather than raw API access. ### [google-tag-manager-mcp-server](https://mcpcatalogs.com/en/server/stape-io-google-tag-manager-mcp-server) by stape-io · score 48 · ★153 MCP server for Google Tag Manager with OAuth authentication, enabling Claude to interact with GTM API. **When to choose:** Choose this server if you need to automate or integrate Google Tag Manager management tasks within Claude workflows and prefer a ready-to-use remote solution with OAuth authentication. **When NOT to choose:** Don't choose this if you require self-hosted solutions, need write access beyond GTM capabilities, or are working with non-Google tag management systems. ### [pluggedin-app](https://mcpcatalogs.com/en/server/veriteknik-pluggedin-app) by VeriTeknik · score 48 · ★94 A unified, self-hostable web interface for discovering, configuring, and managing MCP servers. **When to choose:** Choose plugged.in when you need a centralized hub to manage multiple MCP servers and transform AI conversations into persistent, searchable knowledge. **When NOT to choose:** Don't choose plugged.in if you need a lightweight solution with minimal dependencies or prefer a self-hosted solution without Docker requirements. ### [mcp-ts-core](https://mcpcatalogs.com/en/server/cyanheads-mcp-ts-core) by cyanheads · score 48 · ★138 TypeScript framework for building MCP servers with auth, storage, telemetry, and multi-platform support. **When to choose:** Choose this when building production MCP servers with TypeScript, especially if you need multi-backend storage, authentication, and observability out of the box. **When NOT to choose:** Avoid if you need a simpler solution with fewer dependencies or if you're building a server for environments beyond Bun/Node/Cloudflare Workers. ### [mcp-3D-printer-server](https://mcpcatalogs.com/en/server/dmontgomery40-mcp-3d-printer-server) by DMontgomery40 · score 48 · ★190 MCP server connecting to major 3D printer APIs for print control, monitoring, and STL manipulation. **When to choose:** Choose this server when you need AI integration with 3D printing workflows, especially for Bambu Lab printers where it offers the most comprehensive feature set including advanced STL manipulation and MQTT communication. **When NOT to choose:** Avoid this if you need support for printer systems beyond those listed, require extensive write permissions beyond print control, or are working with extremely large STL files that may exceed memory limitations. ### [unity-api-mcp](https://mcpcatalogs.com/en/server/codeturion-unity-api-mcp) by Codeturion · score 48 · ★56 Unity API MCP server providing instant, accurate Unity documentation lookup for AI agents. **When to choose:** Choose this if you're working with Unity development and want accurate API documentation lookup to prevent hallucinations and reduce token usage in AI-assisted workflows. **When NOT to choose:** Not suitable for commercial use due to PolyForm Noncommercial license limitation; third-party Unity packages aren't covered. ### [mcp-1c](https://mcpcatalogs.com/en/server/feenlace-mcp-1c) by feenlace · score 48 · ★98 MCP server for 1C:Enterprise that provides AI assistants with metadata and generates accurate BSL code through 9 tools. **When to choose:** Choose MCP-1C if you work with 1C:Enterprise and want AI assistants to understand your specific configuration while keeping your data local. **When NOT to choose:** Don't choose MCP-1C if you need access to versions below 8.3.10, or if you require write capabilities (this server is read-only). ### [skrills](https://mcpcatalogs.com/en/server/athola-skrills) by athola · score 48 · ★64 A 36-tool MCP server that validates, analyzes, and syncs skills across Claude Code, Codex CLI, Copilot CLI, and Cursor. **When to choose:** Choose Skrills when working across multiple AI coding assistants (Claude Code, Codex CLI, Copilot CLI, Cursor) and need to maintain skill consistency and validate compatibility across platforms. **When NOT to choose:** Don't choose Skrills if you only use a single AI coding assistant or if you need runtime skill injection into prompts, as Skrills only validates and syncs files without runtime capabilities. ### [ncp](https://mcpcatalogs.com/en/server/portel-dev-ncp) by portel-dev · score 48 · ★90 NCP is an MCP server that unifies multiple MCP tools into a single interface with smart discovery and code execution. **When to choose:** Choose NCP when you're using multiple MCP servers and find your AI assistant getting overwhelmed with tool schemas, response delays, or context exhaustion. **When NOT to choose:** Don't choose NCP if you're only using a small number of MCP tools (2-3) and don't need the unified interface, or if you require direct access to individual MCP tools without abstraction. ### [domscribe](https://mcpcatalogs.com/en/server/patchorbit-domscribe) by patchorbit · score 48 · ★168 Domscribe bridges the gap between running web applications and their source code via MCP, enabling AI agents to view live UI context and make targeted edits. **When to choose:** Choose Domscribe when working with AI coding agents on complex web applications and you need bidirectional communication between the running UI and source code. **When NOT to choose:** Avoid Domscribe if you're working with non-web applications or need to preserve production builds without any instrumentation. ### [mcp-reticle](https://mcpcatalogs.com/en/server/soth-ai-mcp-reticle) by soth-ai · score 48 · ★110 Reticle is a debugging tool for MCP servers that intercepts, visualizes, and profiles JSON-RPC traffic in real-time with zero latency overhead. **When to choose:** Choose Reticle when you need to debug MCP communication issues, profile tool performance, or visualize request-response flows between your LLM and MCP servers. **When NOT to choose:** Avoid Reticle if you need production-level monitoring with alerts, or if you require advanced analytics beyond basic traffic inspection and profiling. ### [django-ai-boost](https://mcpcatalogs.com/en/server/vintasoftware-django-ai-boost) by vintasoftware · score 48 · ★88 Django AI Boost is a production-ready MCP server for Django application introspection and analysis. **When to choose:** Choose Django AI Boost if you work with Django applications and want AI assistants to understand your project structure, models, and configurations. **When NOT to choose:** Don't choose this if you need write access to your Django project or are working with a non-Django Python framework. ### [mcp](https://mcpcatalogs.com/en/server/magicuidesign-mcp) by magicuidesign · score 48 · ★185 MCP server for Magic UI component registry, enabling developers to browse and search UI components. **When to choose:** Choose this MCP server when you're working with UI components and need to integrate Magic UI elements directly into your development workflow through supported IDEs. **When NOT to choose:** Avoid if you need broader UI component libraries beyond Magic UI, as this server is specialized for a specific component ecosystem. ### [OverleafMCP](https://mcpcatalogs.com/en/server/mjyoo2-overleafmcp) by mjyoo2 · score 48 · ★128 An MCP server for AI assistants to read and edit Overleaf LaTeX projects via Git integration. **When to choose:** Choose this MCP server when you need AI assistance with LaTeX document management in Overleaf projects, especially for academic papers or technical documentation requiring section-level edits. **When NOT to choose:** Avoid if you're not using Overleaf, need non-LaTeX document support, or require authentication beyond simple project tokens with no access controls. ### [drawio-mcp-server](https://mcpcatalogs.com/en/server/simonkurtz-msft-drawio-mcp-server) by simonkurtz-MSFT · score 48 · ★71 Draw.io MCP server enabling programmatic diagram generation with 700+ Azure icons and batch operations. **When to choose:** Choose this when you need to generate diagrams programmatically without human intervention, especially when working with Azure architecture or technical documentation. **When NOT to choose:** Don't choose this if you need real-time collaborative editing or export formats beyond XML, as it focuses on programmatic generation rather than interactive design. ### [aicode-toolkit](https://mcpcatalogs.com/en/server/agiflow-aicode-toolkit) by AgiFlow · score 48 · ★156 MCP server toolkit providing project scaffolding, code guidance, and design system discovery for AI agents. **When to choose:** Choose this toolkit when working with multiple coding agents to maintain consistency across projects or when implementing standardized scaffolding and design patterns. **When NOT to choose:** Avoid if you need a simple solution for small projects or if you're concerned about vendor lock-in with the AgiFlow-specific MCP servers. ### [Go-High-Level-MCP-2026-Complete](https://mcpcatalogs.com/en/server/busybee3333-go-high-level-mcp-2026-complete) by BusyBee3333 · score 48 · ★51 GoHighLevel MCP server with 834 tools across 40+ categories for CRM automation, voice AI, and marketing workflows. **When to choose:** Choose this MCP server if you're building agents or workflows that need to interact with GoHighLevel's CRM platform, especially if you want comprehensive API coverage with both raw and curated tools. **When NOT to choose:** Avoid if you're not already using GoHighLevel as this creates vendor lock-in, or if you need tools outside GoHighLevel's scope as this server only exposes GoHighLevel operations. ### [ros2_mcp](https://mcpcatalogs.com/en/server/wise-vision-ros2-mcp) by wise-vision · score 48 · ★77 Python MCP server bridging AI agents with ROS 2 robotics, enabling topic pub/sub, service calls, and action control. **When to choose:** Choose this MCP server when working with ROS 2 robotics and need AI integration for development, debugging, or real-time control without deep ROS expertise. **When NOT to choose:** Not suitable for production environments requiring robust security or for non-ROS 2 robotics systems where the extensive ROS tooling would be unnecessary overhead. ### [zig-mcp](https://mcpcatalogs.com/en/server/zig-wasm-zig-mcp) by zig-wasm · score 48 · ★158 MCP server providing up-to-date documentation for Zig programming language standard library and builtin functions. **When to choose:** Choose this server if you work with Zig and need reliable, up-to-date documentation access through MCP without token-heavy HTML responses. **When NOT to choose:** Don't choose this if you need documentation for languages other than Zig, or if you require write access to documentation (this server is read-only). ### [eclipse-chatgpt-plugin](https://mcpcatalogs.com/en/server/gradusnikov-eclipse-chatgpt-plugin) by gradusnikov · score 48 · ★148 Eclipse IDE plugin exposing development environment as an MCP server for AI agents to code, build, and debug. **When to choose:** Choose this MCP server if you're already using Eclipse for Java development and want to integrate AI agents with your IDE's full functionality while maintaining workspace synchronization and local history. **When NOT to choose:** Don't choose this if you're not using Eclipse or Java development, or if you need cross-IDE support beyond the Eclipse ecosystem. ### [env-doctor](https://mcpcatalogs.com/en/server/mitulgarg-env-doctor) by mitulgarg · score 48 · ★142 Env-Doctor MCP server provides AI assistants with tools to diagnose and fix CUDA/GPU environment compatibility issues. **When to choose:** Choose Env-Doctor when you need to diagnose or fix GPU/CUDA compatibility issues, especially for AI/ML workflows, or when integrating GPU diagnostics into AI assistants via MCP. **When NOT to choose:** Avoid if you only need basic GPU information (it's more complex than necessary), or if you're looking for cross-platform compatibility beyond the listed supported systems. ### [adeu](https://mcpcatalogs.com/en/server/dealfluence-adeu) by dealfluence · score 48 · ★77 Adeu is an MCP server that translates between DOCX and Markdown files, enabling AI to safely edit documents with track changes. **When to choose:** Choose Adeu when you need AI agents to review or edit legal documents while preserving exact formatting and creating proper tracked changes. **When NOT to choose:** Don't choose Adeu if you need to work with document formats other than DOCX or if you prefer a fully open-source solution without cloud dependencies. ### [rust-docs-mcp-server](https://mcpcatalogs.com/en/server/govcraft-rust-docs-mcp-server) by Govcraft · score 48 · ★276 MCP server providing up-to-date Rust crate documentation via semantic search and LLM summarization. **When to choose:** Choose this when you're working with Rust and need up-to-date documentation for specific crates that your AI assistant might not have in its training data. **When NOT to choose:** Avoid this if you don't have access to an OpenAI API key or if you need offline functionality as it requires network access for initial setup and API calls. ### [task-orchestrator](https://mcpcatalogs.com/en/server/jpicklyk-task-orchestrator) by jpicklyk · score 48 · ★183 MCP server providing server-enforced workflow discipline for AI agents with persistent work items, dependency graphs, and actor attribution. **When to choose:** Teams building complex multi-agent workflows that need enforced dependency ordering, quality gates, and audit trails with actor attribution. **When NOT to choose:** Simple workflows without complex dependencies or teams already heavily invested in prompt-based orchestration approaches. ### [code-assistant](https://mcpcatalogs.com/en/server/stippi-code-assistant) by stippi · score 48 · ★163 Rust-based AI coding assistant with GUI, CLI, MCP and ACP modes offering autonomous code analysis and modification. **When to choose:** Choose this when you need a sophisticated AI coding assistant with multiple LLM provider support, especially if you're working on Rust projects or need format-on-save functionality. **When NOT to choose:** Avoid this if you need simple code completion without session state or prefer a more lightweight solution without GUI dependencies. ### [rust-mcp-sdk](https://mcpcatalogs.com/en/server/rust-mcp-stack-rust-mcp-sdk) by rust-mcp-stack · score 48 · ★175 High-performance Rust toolkit for building MCP servers and clients with latest protocol support. **When to choose:** Choose this SDK when building high-performance MCP servers or clients in Rust, especially when you need advanced features like OAuth authentication, multiple transport methods, or latest protocol support. **When NOT to choose:** Not recommended if you're working in a language other than Rust or need a more mature, stable SDK with extensive real-world usage. ### [tmux-mcp](https://mcpcatalogs.com/en/server/nickgnd-tmux-mcp) by nickgnd · score 48 · ★278 A MCP server enabling Claude Desktop to interact with and control tmux terminal sessions. **When to choose:** Choose this MCP server if you want an AI assistant to help manage, analyze, or interact with your tmux terminal sessions from within Claude Desktop. **When NOT to choose:** Avoid this if you need tight security controls over command execution, as the server has full access to your terminal sessions. ### [knowledge-rag](https://mcpcatalogs.com/en/server/lyonzin-knowledge-rag) by lyonzin · score 48 · ★79 Knowledge RAG is a local-first RAG system with 12 MCP tools for document search and retrieval across 20+ file formats. **When to choose:** Choose this for local-first document search when you need privacy, no internet connectivity, or want to avoid API costs. **When NOT to choose:** Avoid this if you need real-time collaborative features or require searching across documents stored in cloud services. ### [unity-mcp-pro-plugin](https://mcpcatalogs.com/en/server/youichi-uda-unity-mcp-pro-plugin) by youichi-uda · score 48 · ★58 Unity MCP Pro plugin provides 280+ AI tools for Unity development via WebSocket MCP connection. **When to choose:** Choose Unity MCP Pro if you're developing games in Unity and want to integrate AI assistants directly into your workflow for automated testing, code generation, and scene manipulation. **When NOT to choose:** Don't choose this if you need AI integration with non-Unity engines, or if you prefer a self-contained solution without the need for a separate MCP server component. ### [Claude-Skills](https://mcpcatalogs.com/en/server/borghei-claude-skills) by borghei · score 48 · ★138 Comprehensive skill library with MCP server integration for Claude Code and other AI assistants. **When to choose:** Choose Claude-Skills when you need a comprehensive library of domain-specific AI skills and tools that can be accessed via MCP server, especially for teams requiring compliance frameworks and industry-specific advisors. **When NOT to choose:** Don't choose Claude-Skills if you need write access capabilities or prefer a more focused toolset without the breadth of multiple domains, as it primarily provides read-only skill access. ### [UnrealMotionGraphicsMCP](https://mcpcatalogs.com/en/server/winyunq-unrealmotiongraphicsmcp) by winyunq · score 48 · ★150 MCP server for Unreal Engine UMG UI automation with blueprint integration and JSON-based workflow. **When to choose:** Choose this MCP server if you're working with Unreal Engine UMG UIs and want AI assistance in designing, modifying, and versioning UI layouts through a Python interface. **When NOT to choose:** Avoid this if you're not using Unreal Engine or don't work with UMG UIs, as it's specialized for this specific workflow. ### [turbo-flow](https://mcpcatalogs.com/en/server/marcuspat-turbo-flow) by marcuspat · score 48 · ★161 Advanced agentic development environment with 215+ MCP tools, GitNexus knowledge graph, and cross-session memory via Beads. **When to choose:** Teams looking for a comprehensive agentic development environment with multi-agent coordination, cross-session memory, and extensive tooling already integrated. **When NOT to choose:** Simple projects or individual developers needing lightweight solutions, as Turbo Flow is a heavy-weight environment with significant setup overhead. ### [mcp-bsl-platform-context](https://mcpcatalogs.com/en/server/alkoleft-mcp-bsl-platform-context) by alkoleft · score 48 · ★165 MCP server providing 1C:Enterprise platform syntax and object model reference for AI assistants. **When to choose:** Choose this MCP server if you're working with 1C:Enterprise development and want AI assistants to have direct access to platform syntax and object model documentation. **When NOT to choose:** Don't choose this if you don't work with 1C:Enterprise platform, as the documentation is specific to this Russian business application platform. ### [erickwendel-contributions-mcp](https://mcpcatalogs.com/en/server/erickwendel-erickwendel-contributions-mcp) by ErickWendel · score 48 · ★232 A TypeScript MCP server providing tools to query Erick Wendel's contributions across different platforms with full MCP capabilities. **When to choose:** Choose this server if you need to query Erick Wendel's technical content contributions through MCP-compatible clients and want to filter by multiple criteria like language, year, or topic. **When NOT to choose:** Don't choose this server if you need access to content from other developers or want a more general-purpose contributions aggregator. ### [goku](https://mcpcatalogs.com/en/server/jcaromiq-goku) by jcaromiq · score 48 · ★145 Goku is a Rust-based HTTP load testing tool with MCP server integration for automated benchmarking. **When to choose:** Choose Goku when you need automated HTTP load testing integrated into AI workflows or CI/CD pipelines, especially when variable templating and multi-step scenarios are required. **When NOT to choose:** Avoid Goku if you need load testing for non-HTTP protocols, as it only supports HTTP/1.1 and HTTP/2. Also consider alternatives if you need database protocol testing or WebSocket support. ### [mcptools](https://mcpcatalogs.com/en/server/posit-dev-mcptools) by posit-dev · score 48 · ★174 mcptools implements the Model Context Protocol in R, enabling R to act as both MCP server and client for AI integration. **When to choose:** Choose mcptools if you're working with R and want AI assistants to interact with your R sessions, access your environment variables, or execute code directly within your R workflow. **When NOT to choose:** Avoid mcptools if you're not working with R, as it's specifically designed for the R ecosystem and won't be useful for other programming languages. ### [boilerplate-mcp-server](https://mcpcatalogs.com/en/server/aashari-boilerplate-mcp-server) by aashari · score 48 · ★70 Production-ready TypeScript MCP server boilerplate with IP lookup tools and layered architecture. **When to choose:** Choose this template when starting a new MCP server project in TypeScript, especially if you need production-ready patterns with security features and token-efficient output formats. **When NOT to choose:** Don't choose this if you need a runtime in JavaScript instead of TypeScript, or if you're looking for a simpler boilerplate without the layered architecture complexity. ### [xc-mcp](https://mcpcatalogs.com/en/server/conorluddy-xc-mcp) by conorluddy · score 48 · ★84 Intelligent Xcode MCP server for iOS development with progressive disclosure to reduce context usage. **When to choose:** Choose XC-MCP if you're working with Xcode and iOS development and need to integrate with AI agents that need to interact with simulators, build systems, and UI elements while being mindful of token limitations. **When NOT to choose:** Avoid XC-MCP if you need direct control over Xcode without token constraints or if you're not working in the Apple ecosystem as it's specifically tailored for iOS/macOS development. ### [Express-REST-API-and-MCP-Server-Framework](https://mcpcatalogs.com/en/server/iolufemi-express-rest-api-and-mcp-server-framework) by iolufemi · score 48 · ★147 Express-based framework that generates dual REST API and MCP servers from a single codebase. **When to choose:** Choose this when you need to build both REST APIs for traditional applications and MCP interfaces for AI clients from the same codebase. **When NOT to choose:** Don't choose this if you need a simple MCP server without REST API requirements, or if you're not using Node.js/Express. ### [mcp-dap-server](https://mcpcatalogs.com/en/server/go-delve-mcp-dap-server) by go-delve · score 48 · ★84 MCP server that bridges AI assistants with DAP debuggers for programmatic debugging of live programs. **When to choose:** Choose this server when you need AI agents to programmatically control debugging sessions across different programming languages with DAP-compatible debuggers. **When NOT to choose:** Don't choose this if you need debugging capabilities for languages without DAP support or if you require a graphical user interface for debugging. ### [regenerator2000](https://mcpcatalogs.com/en/server/ricardoquesada-regenerator2000) by ricardoquesada · score 48 · ★158 Interactive 6502 disassembler with MCP server for AI-assisted reverse engineering. **When to choose:** Choose Regenerator 2000 when working with Commodore 8-bit binaries and need both manual disassembly capabilities and automated analysis through MCP integration. **When NOT to choose:** Don't choose this if you need disassembly support for systems other than Commodore 8-bit computers or require more generalized CPU analysis capabilities. ### [powhttp-mcp](https://mcpcatalogs.com/en/server/usestring-powhttp-mcp) by usestring · score 48 · ★71 MCP server enabling AI assistants to analyze HTTP traffic captured by powhttp with advanced debugging capabilities. **When to choose:** Choose this server when you need to analyze HTTP traffic patterns, debug web scraping issues, or generate scrapers from captured traffic. **When NOT to choose:** Avoid if you don't use powhttp as your traffic capture tool, as this MCP server specifically requires it as a dependency. ### [foundry-mcp-server](https://mcpcatalogs.com/en/server/praneshasp-foundry-mcp-server) by PraneshASP · score 48 · ★247 An MCP server for Solidity developers providing comprehensive Foundry toolchain integration. **When to choose:** Choose this server if you're working with Solidity development and want AI assistants to directly interact with the Foundry ecosystem for contract analysis, testing, and deployment. **When NOT to choose:** Avoid this server if you need blockchain support for networks other than EVM-compatible ones, or if you're concerned about the security implications of providing private keys to AI systems. ### [mcp-server-atlassian-bitbucket](https://mcpcatalogs.com/en/server/aashari-mcp-server-atlassian-bitbucket) by aashari · score 48 · ★146 MCP server connecting AI assistants to Atlassian Bitbucket for repository and PR management. **When to choose:** Choose this MCP server if you're using Bitbucket Cloud and want AI assistants to directly interact with your repositories, pull requests, and codebase through natural language commands. **When NOT to choose:** Don't choose this if you're using Bitbucket Server/Data Center, require write access beyond PR management, or need to integrate with Atlassian's other products like Jira. ### [slither-mcp](https://mcpcatalogs.com/en/server/trailofbits-slither-mcp) by trailofbits · score 48 · ★89 An MCP server wrapping Slither for static analysis of Solidity smart contracts. **When to choose:** Choose this when you need to integrate static security analysis of Solidity contracts into your AI workflows or development process. **When NOT to choose:** Don't choose this if you need real-time analysis or working with non-Solidity smart contract languages. ### [VibeUE](https://mcpcatalogs.com/en/server/kevinpbuckley-vibeue) by kevinpbuckley · score 48 · ★184 VibeUE is an AI-powered Unreal Engine development tool with MCP server integration providing Python API access and specialized tools for game development workflows. **When to choose:** Choose VibeUE when working with Unreal Engine development and wanting AI assistance through natural language commands, especially for terrain generation, asset management, and debugging workflows. **When NOT to choose:** Don't choose VibeUE if you're working on a project that requires offline functionality, or if you prefer open-source solutions without vendor lock-in to VibeUE's API. ### [sverklo](https://mcpcatalogs.com/en/server/sverklo-sverklo) by sverklo · score 48 · ★54 Sverklo is a local-first MCP server providing code intelligence with symbol graphs, blast-radius analysis, and git-pinned memory. **When to choose:** Choose Sverklo for private codebases where code cannot leave the machine, or when you need bi-temporal memory to track design decisions across git history. **When NOT to choose:** Not suitable if you need cloud-based collaboration features or real-time synchronization across multiple developers' workspaces. ### [fieldflow](https://mcpcatalogs.com/en/server/guillaumegay13-fieldflow) by guillaumegay13 · score 47 · ★111 FieldFlow converts OpenAPI endpoints into selectively filtered tools with MCP support for AI clients like Claude Desktop. **When to choose:** Choose FieldFlow when working with OpenAPI-compliant APIs and need precise field selection to minimize token usage in LLM interactions. **When NOT to choose:** Don't choose FieldFlow if you need write operations or direct database access, as it only provides read-only API proxy functionality. ### [mcp-server](https://mcpcatalogs.com/en/server/cap-js-mcp-server) by cap-js · score 47 · ★96 MCP server for AI-assisted development of CAP applications with model and documentation search capabilities. **When to choose:** Choose this if you're developing SAP CAP applications and want AI assistants to understand your project structure and entities. **When NOT to choose:** Don't choose this if you're not working with SAP CAP applications, as it's specialized for that ecosystem only. ### [agent-bridge](https://mcpcatalogs.com/en/server/raysonmeng-agent-bridge) by raysonmeng · score 47 · ★118 Bidirectional bridge between Claude Code and Codex via MCP protocol for collaborative workflows. **When to choose:** Choose this if you're using both Claude Code and Codex and want seamless bidirectional collaboration between them in the same workflow. **When NOT to choose:** Not suitable if you're not using both Claude Code and Codex, or if you need multi-session or multi-project support, which isn't available yet. ### [api2mcp4j](https://mcpcatalogs.com/en/server/theeterna-api2mcp4j) by TheEterna · score 47 · ★102 Spring Boot starter that automatically converts REST APIs to MCP servers with zero to low code. **When to choose:** Choose api2mcp4j when you have existing Spring Boot REST APIs that you want to expose as MCP tools without refactoring your business logic. **When NOT to choose:** Don't choose this if you need a stable, published artifact (currently only available as a build from source) or if you're not using Spring Boot. ### [mcpc](https://mcpcatalogs.com/en/server/mcpc-tech-mcpc) by mcpc-tech · score 47 · ★93 MCPC is an SDK for building agentic MCP servers by composing existing MCP tools. **When to choose:** Choose MCPC when you need to build sophisticated multi-agent systems that leverage existing MCP tools and require flexible composition, fine-tuning capabilities, and multiple execution modes. **When NOT to choose:** Avoid MCPC if you need a simple MCP tool without composition capabilities or if you're looking for a turnkey solution with minimal development overhead. ### [advanced-unity-mcp](https://mcpcatalogs.com/en/server/codemaestroai-advanced-unity-mcp) by codemaestroai · score 47 · ★91 An MCP server that enables AI assistants to control Unity Editor through natural language commands. **When to choose:** Choose this server if you're working with Unity and want to control the editor through natural language commands, especially if you're already using Code Maestro or other supported MCP clients. **When NOT to choose:** Don't choose this if you're using a Unity version below 2022, or if you need to use unsupported MCP clients without a relay server connection. ### [MCPSharp](https://mcpcatalogs.com/en/server/afrise-mcpsharp) by afrise · score 47 · ★369 MCPSharp is a .NET library for building MCP servers and clients with tools, resources, and prompts. **When to choose:** Teams working in .NET environments who need to expose existing .NET code as MCP tools without implementing protocol details. **When NOT to choose:** If you're not working in .NET ecosystems or need advanced MCP features not yet supported by MCPSharp. ### [mcp-superset](https://mcpcatalogs.com/en/server/bintocher-mcp-superset) by bintocher · score 47 · ★36 A comprehensive MCP server with 137 tools for managing Apache Superset dashboards, charts, datasets, and security features. **When to choose:** Choose mcp-superset when you need comprehensive control over Apache Superset through AI assistants, particularly for security management, permissions auditing, and dashboard automation that other Superset MCP servers don't provide. **When NOT to choose:** Don't choose this if you're using an older version of Superset (pre-4.1.1), need LDAP authentication as the only option, or require a lightweight solution with minimal tools as the 137-tool coverage may be overwhelming for simple use cases. ### [mcp-use-ts](https://mcpcatalogs.com/en/server/mcp-use-mcp-use-ts) by mcp-use · score 47 · ★174 Complete TypeScript framework for MCP with client/server SDKs, React hooks, and built-in debugging tools. **When to choose:** Choose MCP-Use when you're building TypeScript-based MCP applications that require rich UI components, debugging capabilities, and a comprehensive development framework. **When NOT to choose:** Avoid if you need minimal dependencies or are building a simple MCP server without UI components, as MCP-Use provides a full-featured framework with more overhead. ### [godoc-mcp](https://mcpcatalogs.com/en/server/mrjoshuak-godoc-mcp) by mrjoshuak · score 47 · ★117 MCP server that provides efficient access to Go documentation to reduce token usage when AI coding agents work with Go projects. **When to choose:** Choose godoc-mcp when working with Go projects and need efficient token usage when documenting code, especially with large codebases where reading entire source files would be too expensive. **When NOT to choose:** Don't use godoc-mcp if you're primarily working with languages other than Go, or if you need direct access to source code rather than documentation. ### [Embody](https://mcpcatalogs.com/en/server/dylanroscover-embody) by dylanroscover · score 47 · ★102 MCP server for TouchDesigner that lets AI assistants build, wire, and debug networks with natural language commands. **When to choose:** Choose Embody if you work with TouchDesigner and want AI assistance to build, wire, and debug networks directly through MCP protocol, plus version control for your creative coding projects. **When NOT to choose:** Don't choose Embody if you don't use TouchDesigner, as it's tightly integrated with this specific creative coding platform. ### [mcp-rest-api](https://mcpcatalogs.com/en/server/dkmaker-mcp-rest-api) by dkmaker · score 47 · ★95 TypeScript MCP server for testing REST APIs through MCP clients like Cline. **When to choose:** Choose this tool when you need to directly test REST APIs from your development environment using various MCP clients, especially if you need support for multiple authentication methods. **When NOT to choose:** Avoid this if you need advanced API testing features like test suites, mocking, or performance testing as it focuses on basic endpoint testing functionality. ### [mcpadapt](https://mcpcatalogs.com/en/server/grll-mcpadapt) by grll · score 47 · ★420 MCPAdapt bridges 650+ MCP servers to various agentic frameworks like LangChain and SmolAgents. **When to choose:** Choose MCPAdapt when you need to integrate MCP servers into existing agent frameworks like Langchain or Smolagents, especially when working with multiple MCP servers. **When NOT to choose:** Don't choose MCPAdapt if you need direct MCP server implementation without framework adapters or if you require authentication mechanisms beyond basic SSE support. ### [mcp-gitlab-server](https://mcpcatalogs.com/en/server/yoda-digital-mcp-gitlab-server) by yoda-digital · score 47 · ★49 Production-ready GitLab MCP server with 86 tools, full API coverage, multiple auth modes and transports. **When to choose:** Choose this for comprehensive GitLab integration with AI agents when you need extensive coverage of GitLab features (86 tools) and production-ready deployments with proper authentication. **When NOT to choose:** Don't choose if you only need basic GitLab operations or if you're already using GitLab's built-in MCP server which provides 15 tools with native GitLab authentication. ### [mcp-forge](https://mcpcatalogs.com/en/server/achetronic-mcp-forge) by achetronic · score 47 · ★90 Production-ready MCP server template in Go with OAuth authentication and deployment examples. **When to choose:** Choose this template when building a production MCP server in Go that requires OAuth authentication and needs to support both local and remote AI providers. **When NOT to choose:** Don't choose this if you need a server in a language other than Go, or if you require simpler authentication without OAuth complexity. ### [Delphi-MCP-Server](https://mcpcatalogs.com/en/server/gdksoftware-delphi-mcp-server) by GDKsoftware · score 47 · ★117 A native Delphi implementation of MCP protocol with HTTP and STDIO transport modes for AI-powered Delphi development workflows. **When to choose:** Choose this when working in Delphi environments and need native MCP protocol implementation without external dependencies. **When NOT to choose:** Not suitable if you're not using Delphi, as this is specifically implemented for the Delphi ecosystem. ### [aai-gateway](https://mcpcatalogs.com/en/server/gybob-aai-gateway) by gybob · score 47 · ★104 AAI Gateway provides unified management of MCP servers and skills across multiple AI agents with 99% context token savings. **When to choose:** Choose AAI Gateway when you need to manage multiple MCP servers across different AI agents and want to reduce context overhead significantly. **When NOT to choose:** Not ideal if you need fine-grained control over individual MCP configurations or are working with an AI agent not yet supported by AAI Gateway. ### [algorand-mcp](https://mcpcatalogs.com/en/server/goplausible-algorand-mcp) by GoPlausible · score 47 · ★45 Algorand MCP server provides blockchain tools with secure wallet management for AI agents. **When to choose:** Choose this server when building AI applications that need secure access to Algorand blockchain operations and DeFi protocols. **When NOT to choose:** Don't choose this if you need support for other blockchains or require custom transaction logic not covered by the existing tools. ### [mcp-server-spec-driven-development](https://mcpcatalogs.com/en/server/formulahendry-mcp-server-spec-driven-development) by formulahendry · score 47 · ★431 MCP server enabling structured spec-driven development with requirements, design, and code generation prompts. **When to choose:** Choose this MCP server when you want to implement a structured, specification-driven development approach that ensures traceability from requirements through design to implementation. **When NOT to choose:** Don't choose this if you prefer an agile, iterative approach without extensive documentation or if you're working on small, quick projects that don't benefit from formal specifications. ### [claude-emporium](https://mcpcatalogs.com/en/server/vvkmnn-claude-emporium) by Vvkmnn · score 47 · ★104 A collection of 6 MCP plugins that wrap standalone servers with automation hooks, commands, and skills. **When to choose:** Choose this suite if you're building complex Claude workflows that need persistent context, session memory, or automated error handling with a Roman-themed aesthetic. **When NOT to choose:** Avoid if you prefer minimal MCP setups without additional abstractions or if you're not comfortable with the theme-based approach to server organization. ### [mcp-graphql](https://mcpcatalogs.com/en/server/blurrah-mcp-graphql) by blurrah · score 47 · ★385 MCP server enabling LLMs to interact with GraphQL APIs through schema introspection and query execution. **When to choose:** Choose mcp-graphql when you need to connect LLMs to GraphQL APIs and want schema introspection capabilities with controlled query access. **When NOT to choose:** Don't choose this if you need advanced authentication mechanisms beyond basic headers, or if you're working with non-GraphQL REST APIs. ### [photon](https://mcpcatalogs.com/en/server/portel-dev-photon) by portel-dev · score 47 · ★94 Photon turns TypeScript files into MCP servers, CLI tools, and web interfaces with minimal boilerplate. **When to choose:** Choose Photon when you need to create tools that will be used by both humans (via web/CLI) and AI agents (via MCP) from a single codebase definition. **When NOT to choose:** Avoid Photon if you need fine-grained control over MCP server implementation details or if you're already invested in a different MCP framework that meets your needs. ### [chat.md](https://mcpcatalogs.com/en/server/rusiaaman-chat-md) by rusiaaman · score 47 · ★105 A VS Code extension that turns .chat.md files into editable AI chat interfaces with MCP support. **When to choose:** Choose this when you want editable, version-controlled AI conversations stored as regular files in your workspace, especially if you work with multiple LLM providers. **When NOT to choose:** Don't choose this if you need real-time voice conversations or if you prefer web-based AI interfaces over file-based solutions. ### [automation-mcp](https://mcpcatalogs.com/en/server/ashwwwin-automation-mcp) by ashwwwin · score 47 · ★391 A MCP server for macOS desktop automation with mouse, keyboard, screen capture, and window management. **When to choose:** Choose this server when you need comprehensive desktop automation on macOS, especially for UI testing, task automation, or building accessibility tools that require precise control over mouse, keyboard, and windows. **When NOT to choose:** Don't choose this if you need cross-platform automation, as it only works on macOS. Also consider alternatives if you require more complex image recognition capabilities beyond simple template matching. ### [mcp-dock](https://mcpcatalogs.com/en/server/oldjii-mcp-dock) by OldJii · score 47 · ★122 Cross-platform MCP server manager for popular AI clients with one-click installation and multi-client sync. **When to choose:** Choose MCP Dock if you use multiple AI clients (Cursor, Claude, Windsurf, etc.) and want a unified way to manage MCP servers and configurations across all of them. **When NOT to choose:** Don't choose MCP Dock if you only use a single AI client and prefer direct MCP server management without the overhead of an additional management tool. ### [reshapr](https://mcpcatalogs.com/en/server/reshaprio-reshapr) by reshaprio · score 47 · ★92 reShapr is a no-code MCP server that transforms REST/GraphQL/gRPC APIs into AI-native tools with context control. **When to choose:** Choose reShapr when you need to transform existing REST/GraphQL/gRPC APIs into MCP tools without writing code and want fine-grained control over the context sent to LLMs. **When NOT to choose:** Don't choose reShapr if you need direct write access to APIs (currently read-only), require real-time data streaming capabilities, or need a solution with enterprise-grade security features. ### [fhir-mcp-server](https://mcpcatalogs.com/en/server/wso2-fhir-mcp-server) by wso2 · score 47 · ★119 FHIR MCP Server bridges healthcare FHIR APIs with AI tools via MCP protocol, enabling secure clinical data access. **When to choose:** Choose this server when you need to integrate healthcare data from FHIR-compatible systems into AI applications, especially when working with existing healthcare infrastructure that uses SMART-on-FHIR authentication. **When NOT to choose:** Don't choose this if you're working with non-healthcare data or if your environment doesn't have access to a FHIR server with proper authentication credentials. ### [ida-headless-mcp](https://mcpcatalogs.com/en/server/zboralski-ida-headless-mcp) by zboralski · score 47 · ★115 A headless IDA Pro binary analysis server providing 52 MCP tools for reverse engineering with multi-session concurrency. **When to choose:** Choose this when you need interactive reverse engineering capabilities in AI models, especially for concurrent analysis of multiple binaries with IDA Pro. **When NOT to choose:** Avoid if you don't have an IDA Pro license or need to analyze binaries without commercial tools - this requires IDA Pro 9.0+ as a prerequisite. ### [prism-mcp-rs](https://mcpcatalogs.com/en/server/prismworks-ai-prism-mcp-rs) by prismworks-ai · score 47 · ★44 Enterprise-grade Rust implementation of MCP protocol with production controls and transport flexibility. **When to choose:** Choose prism-mcp-rs when building enterprise-grade MCP systems in Rust requiring production reliability, resilience patterns, and multi-transport support. **When NOT to choose:** Not ideal for teams using other languages (Rust-specific) or for simple MCP implementations without need for enterprise-grade resilience features. ### [gograph](https://mcpcatalogs.com/en/server/ozgurcd-gograph) by ozgurcd · score 47 · ★118 A local AST/type-aware Go repository context indexer that creates optimized graphs for AI coding agents. **When to choose:** Choose gograph when working with Go codebases that need efficient AST analysis and context generation for AI coding assistants. **When NOT to choose:** Don't choose gograph if you're not working with Go codebases or if you specifically need an MCP server implementation. ### [Fusion-360-MCP-Server](https://mcpcatalogs.com/en/server/aurafriday-fusion-360-mcp-server) by AuraFriday · score 47 · ★94 A fully functional MCP server that enables AI agents to control Autodesk Fusion 360 through Python execution, generic API calls, and cross-tool integration. **When to choose:** Choose this when you need AI to control CAD design processes in Fusion 360 with full API access, especially for parametric modeling and mechanical design workflows. **When NOT to choose:** Don't choose this if you're not using Autodesk Fusion 360, as it's specifically designed for this CAD platform and requires it to function. ### [p4mcp-server](https://mcpcatalogs.com/en/server/perforce-p4mcp-server) by perforce · score 47 · ★92 Perforce P4 MCP Server integrates with P4 version control to provide structured read/write tools for changelists, files, and more. **When to choose:** Choose this MCP server if your team already uses Perforce P4 and wants AI assistants to safely interact with version control operations. **When NOT to choose:** Avoid this if you need support for version control systems other than Perforce, or if you require write operations without additional confirmation steps. ### [mcp-nodejs-debugger](https://mcpcatalogs.com/en/server/workbackai-mcp-nodejs-debugger) by workbackai · score 47 · ★300 MCP server enabling runtime debugging of Node.js applications for Cursor and Claude Code. **When to choose:** Choose this tool when working with Node.js applications and debugging directly within your AI coding assistant without switching contexts. **When NOT to choose:** Not suitable for debugging applications in production environments or when working with non-Node.js runtimes. ### [gm-exec](https://mcpcatalogs.com/en/server/anentrypoint-gm-exec) by AnEntrypoint · score 47 · ★122 MCP server for executing code in multiple languages with background processing. **When to choose:** Choose gm-exec when you need secure, multi-language code execution within MCP clients with background task capabilities. **When NOT to choose:** Don't choose gm-exec if you need longer execution times than 15 seconds or require language runtimes not in the supported list. ### [wanaku](https://mcpcatalogs.com/en/server/wanaku-ai-wanaku) by wanaku-ai · score 47 · ★112 Wanaku MCP Router is a Java-based gateway that connects and routes MCP servers with extensive connectivity options. **When to choose:** Choose Wanaku when you need to manage multiple MCP servers in enterprise environments, require secure routing with authentication, and need Kubernetes-native deployment. **When NOT to choose:** Don't choose Wanaku if you need a simple MCP server for basic tool integration or if you're looking for a lightweight solution with minimal dependencies. ### [leetcode-mcp-server](https://mcpcatalogs.com/en/server/jinzcdev-leetcode-mcp-server) by jinzcdev · score 47 · ★113 MCP server providing automated access to LeetCode problems, solutions, and user data with optional authentication. **When to choose:** Choose this server if you're a developer preparing for coding interviews or practicing algorithms who wants automated access to LeetCode's problem database and solution analysis. **When NOT to choose:** Don't choose this if you need access to premium LeetCode content or features not exposed through their API, or if you're concerned about sharing your LeetCode credentials. ### [mcp-claude-code](https://mcpcatalogs.com/en/server/sdglbl-mcp-claude-code) by SDGLBL · score 47 · ★301 MCP server implementing Claude Code capabilities with file operations, code analysis, and command execution tools. **When to choose:** Choose this when you need Claude Code-like functionality with enhanced file operations, code search, and multi-agent capabilities for development workflows. **When NOT to choose:** Avoid if you need fine-grained control over individual tool permissions, as this server operates with directory-level access controls. ### [monolith](https://mcpcatalogs.com/en/server/tumourlove-monolith) by tumourlove · score 47 · ★110 Unreal Engine MCP plugin with 1,226 actions across 16 modules, enabling AI assistants to manipulate Blueprints, Materials, Animation, and more. **When to choose:** Choose Monolith when you're working with Unreal Engine projects and need AI assistance with deep manipulation of game assets, Blueprints, materials, animations, and other UE-specific content without Python dependencies. **When NOT to choose:** Don't choose Monolith if you're not working with Unreal Engine, or if you need access to engine versions prior to 5.7. Also reconsider if your project requires features not covered by the 16 built-in modules. ### [mcp-server](https://mcpcatalogs.com/en/server/bitdive-mcp-server) by bitDive · score 47 · ★86 MCP server providing trace analysis, request reproduction, and regression management for AI agents. **When to choose:** Choose this server if you're working with Java applications and need AI agents to analyze runtime traces, reproduce issues, and compare behavior changes without leaving your development environment. **When NOT to choose:** Avoid this server if you're not using BitDive for monitoring or if you need support for non-Java applications, as it's tightly integrated with the BitDive ecosystem. ### [claudex](https://mcpcatalogs.com/en/server/kunwar-shah-claudex) by kunwar-shah · score 47 · ★88 Claudex is an MCP server that provides persistent memory and FTS5 search for Claude Code conversation history with a web UI. **When to choose:** Choose Claudex when you need to analyze and search Claude Code conversation history while providing persistent memory to Claude Code across sessions. **When NOT to choose:** Don't choose Claudex if you're using a different AI assistant that doesn't integrate with Claude Code, or if you need a solution for conversation analysis that doesn't require MCP integration. ### [code-context-engine](https://mcpcatalogs.com/en/server/elara-labs-code-context-engine) by elara-labs · score 47 · ★120 Local MCP server that indexes codebases to reduce AI token consumption by 94% through semantic search and compression. **When to choose:** Choose this if you're using Claude Code or other AI coding tools and want significant token savings while keeping all code indexing private and local. **When NOT to choose:** Avoid if you need write access to your repository (this is read-only) or prefer cloud-based solutions that offer collaborative features beyond token reduction. ### [mcp-gopls](https://mcpcatalogs.com/en/server/hloiseau-mcp-gopls) by hloiseau · score 47 · ★84 MCP server for Go using gopls – LSP-powered analysis, tests, coverage, and tooling. **When to choose:** Choose mcp-gopls when you need comprehensive Go language tooling through MCP, including LSP features not available in gopls's built-in MCP server, and you're already working with Go projects. **When NOT to choose:** Avoid if you only need basic introspection tools like diagnostics and symbol search, in which case gopls's built-in MCP server would be sufficient without requiring additional binaries. ### [CAD-MCP](https://mcpcatalogs.com/en/server/daobataotie-cad-mcp) by daobataotie · score 47 · ★338 MCP server that controls CAD software through natural language commands for drawing operations. **When to choose:** Choose CAD-MCP when you need to control CAD software through natural language commands and are already using Windows with supported CAD applications like AutoCAD. **When NOT to choose:** Don't choose CAD-MCP if you need cross-platform support, work exclusively with CAD software not on the supported list, or require advanced CAD automation beyond drawing basic shapes and saving files. ### [mcp-codebase-index](https://mcpcatalogs.com/en/server/mikerecognex-mcp-codebase-index) by MikeRecognex · score 47 · ★51 Zero-dependency MCP server indexing 17 tools for codebase navigation with persistent caching and incremental updates. **When to choose:** Developers working with medium-to-large Python/TypeScript/Go/Rust/C# codebases who need efficient code navigation and impact analysis without paying token costs for entire file reads. **When NOT to choose:** Projects using languages not supported by this indexer, or when you need write access to codebases (this tool is read-only). ### [native-devtools-mcp](https://mcpcatalogs.com/en/server/sh3ll3x3c-native-devtools-mcp) by sh3ll3x3c · score 47 · ★99 Multi-platform MCP server for desktop automation with screenshots, OCR, input simulation, browser automation via CDP, and Android support. **When to choose:** Choose this when you need cross-platform automation (macOS + Windows + Android) in a single MCP server, especially when element-precise interaction with native macOS apps without moving the cursor is required. **When NOT to choose:** Don't choose this if you need Linux support, iOS automation, or are working exclusively with Firefox/WebKit browsers which aren't supported. ### [Claude-Patent-Creator](https://mcpcatalogs.com/en/server/robthepcguy-claude-patent-creator) by RobThePCGuy · score 47 · ★118 An AI-powered patent creation and analysis MCP server with hybrid RAG search over patent regulations, BigQuery access to 76M+ patents, and automated compliance checking. **When to choose:** Choose this if you're working with US patent creation and need automated compliance checking, regulation search, and prior art analysis capabilities through MCP. **When NOT to choose:** Don't choose this if you need support for non-US patent systems, as it's currently focused only on USPTO workflows and lacks international patent coverage. ### [blockbench-mcp-plugin](https://mcpcatalogs.com/en/server/jasonjgardner-blockbench-mcp-plugin) by jasonjgardner · score 47 · ★117 A TypeScript plugin that adds MCP server functionality to Blockbench for 3D model editing. **When to choose:** Choose this when you're working with Blockbench for 3D modeling and want AI assistance in your workflow without leaving the application. **When NOT to choose:** Don't choose this if you're not using Blockbench for 3D modeling, as this MCP server only integrates with that specific platform. ### [spring-ai-playground](https://mcpcatalogs.com/en/server/spring-ai-community-spring-ai-playground) by spring-ai-community · score 47 · ★113 Cross-platform desktop app for building, testing, and publishing MCP tools with a 'no-pass-no-run' safety workflow. **When to choose:** Choose Spring AI Playground when you need to build and validate MCP tools without Java expertise, especially when tool safety and validation before agent execution is a priority. **When NOT to choose:** Avoid if you need tool development with full Java/Spring capabilities or if you require a cloud-based tool development environment. ### [mcp-server](https://mcpcatalogs.com/en/server/ui5-mcp-server) by UI5 · score 47 · ★85 The UI5 MCP server provides specialized tools for UI5 application development and integration with AI agents. **When to choose:** Choose this MCP server when working with UI5/SAPUI5 development and want AI assistants to have access to UI5-specific tools for scaffolding, validation, and documentation. **When NOT to choose:** Not suitable if you're not using the UI5 framework or if you need write access capabilities (this server is focused on read-only assistance and scaffolding). ### [rust-docs-mcp](https://mcpcatalogs.com/en/server/snowmead-rust-docs-mcp) by snowmead · score 47 · ★124 A comprehensive MCP server providing Rust crate documentation, source code analysis, dependency trees, and module structure for AI agents. **When to choose:** Choose this MCP server when your AI agent needs comprehensive Rust crate analysis, documentation access, and dependency tracing for building Rust applications. **When NOT to choose:** Don't choose this if you need to work with non-Rust languages or require write access to crates (this server is read-only). ### [mxcp](https://mcpcatalogs.com/en/server/raw-labs-mxcp) by raw-labs · score 47 · ★67 MXCP is an enterprise-grade framework for building production-ready MCP servers with data modeling, security policies, and comprehensive testing. **When to choose:** Choose MXCP when building production-grade AI applications that need enterprise security, compliance auditing, and robust data quality validation. **When NOT to choose:** Avoid MXCP for simple or experimental AI tools as it introduces significant complexity with its enterprise features and BSL license. ### [mcp-anything](https://mcpcatalogs.com/en/server/type-mcp-mcp-anything) by Type-MCP · score 47 · ★36 MCP-Anything generates production-ready MCP servers from briefs, codebases, or API specs in seconds. **When to choose:** Teams that need to quickly convert existing APIs or codebases into MCP servers without manual implementation, especially when working across multiple ecosystems. **When NOT to choose:** When you need fine-grained control over tool design, custom authentication models, or have strict requirements around performance that can't be met by the generated wrapper. ### [gdai-mcp-plugin-godot](https://mcpcatalogs.com/en/server/3ddelano-gdai-mcp-plugin-godot) by 3ddelano · score 47 · ★85 GDAI MCP enables AI assistants to control Godot Engine, automating scene creation, debugging, and game development workflows. **When to choose:** Choose this server when working with Godot Engine game development and want AI assistance to automate scene creation, debug scripts, and manage game assets. **When NOT to choose:** Avoid this if you're not using Godot Engine or need game development capabilities for other engines. ### [action_mcp](https://mcpcatalogs.com/en/server/seuros-action-mcp) by seuros · score 47 · ★113 A Ruby gem that provides MCP server capabilities to Rails applications with robust production-ready network-based deployments. **When to choose:** Choose ActionMCP if you're building Rails applications that need to expose tools, resources, and prompts to AI systems through MCP and want a production-ready solution without implementing MCP from scratch. **When NOT to choose:** Don't choose ActionMCP if you need STDIO transport for local development, require a non-Ruby solution, or need compatibility with MCP protocol versions prior to 2025-03-26. ### [template-repo](https://mcpcatalogs.com/en/server/andrewaltimit-template-repo) by AndrewAltimit · score 47 · ★127 Advanced agent orchestration template integrating 20 MCP servers for security, code quality, and multimedia tools. **When to choose:** Choose this template if you're building secure multi-agent systems and need a reference architecture with built-in trust measurement, safety protocols, and extensive MCP server integration. **When NOT to choose:** Don't choose this if you need a simple MCP server for basic tool integration, as this is a complex orchestration template designed for advanced autonomous agent development with significant setup complexity. ### [ophis](https://mcpcatalogs.com/en/server/njayp-ophis) by njayp · score 47 · ★85 Ophis transforms Cobra CLI commands into MCP servers, enabling Claude Desktop, VSCode, and Cursor integration. **When to choose:** Choose Ophis when you have an existing Cobra-based CLI tool that you want to expose as MCP resources without rewriting it as a native MCP server. **When NOT to choose:** Don't choose Ophis if you're starting a new project from scratch, as there are more direct MCP implementations; also avoid if your CLI has complex state management that subprocess execution can't properly handle. ### [CTX](https://mcpcatalogs.com/en/server/alegau03-ctx) by Alegau03 · score 47 · ★122 CTX is a context runtime engine that optimizes AI coding workflows by reducing token waste through graph memory, compact task packs, and caching. **When to choose:** Choose CTX if you're using OpenCode and need significant token reduction through intelligent context management, especially with complex rule sets or repeated file reads. **When NOT to choose:** Don't choose CTX if you're not using OpenCode, or if you prefer traditional markdown-based instruction files over graph-based memory systems. ### [forge-orchestrator](https://mcpcatalogs.com/en/server/nxtg-ai-forge-orchestrator) by nxtg-ai · score 47 · ★116 Rust-based MCP server for multi-AI task orchestration with file locking, knowledge capture and drift detection. **When to choose:** Teams using multiple AI coding tools on the same repository who need coordination, file locking, and knowledge preservation between tool sessions. **When NOT to choose:** If you're only using a single AI coding tool, or need a fully open-source solution (this uses a functional source license that converts to Apache 2.0 in 2028). ### [MCP-Packet-Tracer](https://mcpcatalogs.com/en/server/mats2208-mcp-packet-tracer) by Mats2208 · score 47 · ★40 AI-powered Cisco Packet Tracer automation via MCP with 30 tools for network topology generation, validation and deployment. **When to choose:** Choose this MCP server for network automation education and testing when you need to generate complex Cisco network topologies from natural language descriptions and deploy them in real-time to Packet Tracer. **When NOT to choose:** Don't choose this if you need production network deployment (it only works with Packet Tracer simulation), if you require extensive vendor-specific configurations not covered, or if you need authentication/authorization features. ### [itsyconnect-macos](https://mcpcatalogs.com/en/server/nickustinov-itsyconnect-macos) by nickustinov · score 47 · ★96 App Store Connect alternative desktop app with optional MCP server for managing iOS app listings. **When to choose:** Choose this MCP server if you're developing macOS apps and want to manage App Store Connect listings directly through AI coding tools while maintaining local data storage. **When NOT to choose:** Don't choose this if you need cloud-based team collaboration features or if you're not developing for the Apple ecosystem. ### [gitlab-mr-mcp](https://mcpcatalogs.com/en/server/kopfrechner-gitlab-mr-mcp) by kopfrechner · score 47 · ★89 GitLab MR MCP server allows AI agents to interact with GitLab repositories to manage merge requests and issues. **When to choose:** Choose this server when you need AI agents to interact with GitLab repositories to manage merge requests and issues, particularly if you're already using GitLab as your primary code collaboration platform. **When NOT to choose:** Don't choose this if you need deep integration with GitHub repositories instead of GitLab, or if you require extensive project management features beyond merge requests and issues. ### [In-Memoria](https://mcpcatalogs.com/en/server/pi22by7-in-memoria) by pi22by7 · score 47 · ★168 MCP server that learns code patterns and provides persistent memory for AI assistants. **When to choose:** Choose In-Memoria when you need an AI assistant that remembers your project-specific patterns and conventions across coding sessions, especially when working with Claude or GitHub Copilot. **When NOT to choose:** Don't choose In-Memoria if you need cloud-based AI assistance with multi-project support, as it's designed for local-only, single-project codebases. ### [mcp-neovim-server](https://mcpcatalogs.com/en/server/bigcodegen-mcp-neovim-server) by bigcodegen · score 47 · ★312 MCP server for controlling Neovim with 19 tools for editing, navigation, buffer management and workflow automation. **When to choose:** Choose this MCP server if you're already using Neovim and want AI assistance directly within your text editor environment. **When NOT to choose:** Don't choose this if you use other text editors (VS Code, Emacs, etc.) or if you need complex Neovim plugin integration that might interfere with the MCP server. ### [lastsaas](https://mcpcatalogs.com/en/server/jonradoff-lastsaas) by jonradoff · score 47 · ★97 Go-based SaaS boilerplate with MCP server for AI-powered admin operations. **When to choose:** Choose this MCP server if you're already using or planning to implement LastSaaS as your SaaS foundation and want AI-powered monitoring and analytics capabilities without building custom integrations. **When NOT to choose:** Don't choose this if you need write access to your systems through MCP, or if you're not using LastSaaS as your SaaS foundation since the MCP server is tightly integrated with this specific platform. ### [opencode-mcp](https://mcpcatalogs.com/en/server/alaeddinemessadi-opencode-mcp) by AlaeddineMessadi · score 47 · ★39 MCP server that integrates OpenCode AI's 79 tools with MCP clients for autonomous coding tasks. **When to choose:** Choose opencode-mcp when you need to delegate complex coding tasks to an autonomous AI agent that can read, write, and execute code in your project while working alongside your primary AI assistant. **When NOT to choose:** Don't choose this if you're already invested in a different coding assistant ecosystem or if you prefer direct control over all code changes without delegation to an external tool. ### [codebadger](https://mcpcatalogs.com/en/server/lekssays-codebadger) by Lekssays · score 47 · ★102 Containerized MCP server for static code analysis using Joern's CPG with multi-language support. **When to choose:** Choose CodeBadger when you need deep program analysis capabilities for security auditing, vulnerability hunting, or understanding complex codebases across multiple programming languages. **When NOT to choose:** Avoid CodeBadger if you need real-time analysis, don't want to manage Docker containers, or require support for languages not in its 12-language list. ### [mcp-jenkins](https://mcpcatalogs.com/en/server/lanbaoshen-mcp-jenkins) by lanbaoshen · score 47 · ★120 MCP server for Jenkins integration, providing tools to interact with Jenkins CI/CD through AI assistants. **When to choose:** Choose this server when your team uses Jenkins for CI/CD and wants AI assistants to manage or query pipeline status, build jobs, and Jenkins resources through natural language. **When NOT to choose:** Don't choose this if you use a different CI/CD system (like GitLab CI, GitHub Actions, or Azure DevOps) as this server is specifically designed for Jenkins integration only. ### [better-code-review-graph](https://mcpcatalogs.com/en/server/n24q02m-better-code-review-graph) by n24q02m · score 47 · ★46 MCP server for token-efficient code reviews using knowledge graphs with fixed search and configurable embeddings. **When to choose:** Choose this for AI code review when working with medium to large codebases where token efficiency matters, especially if you need temporal analysis of changes or security scanning. **When NOT to choose:** Don't choose this for very small projects where the overhead of building a knowledge graph outweighs benefits, or if you need real-time collaborative editing features. ### [mcp-typescript-sdk](https://mcpcatalogs.com/en/server/emqx-mcp-typescript-sdk) by emqx · score 47 · ★97 A TypeScript SDK for implementing MCP over MQTT with full type safety, supporting both browser and Node.js environments. **When to choose:** Choose this SDK when building MCP applications that need to communicate between browser and Node.js environments, especially when you already use MQTT infrastructure or need reliable message queuing. **When NOT to choose:** Avoid this if you need direct HTTP/WebSocket MCP transport without MQTT dependency, or if your environment doesn't support MQTT brokers. ### [openmcp](https://mcpcatalogs.com/en/server/getdatanaut-openmcp) by getdatanaut · score 47 · ★306 Turn any OpenAPI specification into an MCP server with just the tools you need. **When to choose:** Choose OpenMCP when you need to quickly convert existing OpenAPI documentation into MCP tools without manual implementation, or when you want to combine multiple API services into a single focused MCP server. **When NOT to choose:** Avoid if you need complex API transformations beyond what OpenAPI supports, or if you require advanced routing/filtering capabilities not provided by the configuration model. ### [ue5-mcp-bridge](https://mcpcatalogs.com/en/server/natfii-ue5-mcp-bridge) by Natfii · score 47 · ★45 MCP server bridging AI assistants to Unreal Engine 5 editor, enabling direct manipulation of game development assets and workflows. **When to choose:** Choose this MCP server if you're already using Unreal Engine 5 and want to accelerate development workflow through AI assistants that can directly manipulate your project assets, Blueprints, and editor functionality. **When NOT to choose:** Don't choose this if you need to work with other game engines, require 3D model/texture generation capabilities, or don't have Unreal Engine 5 installed and configured with the required HTTP backend. ### [emcee](https://mcpcatalogs.com/en/server/mattt-emcee) by mattt · score 47 · ★321 emcee converts OpenAPI specifications to MCP servers, enabling Claude Desktop and other apps to connect to web APIs. **When to choose:** Choose emcee when you need to quickly connect Claude Desktop to existing APIs with OpenAPI specifications without writing custom MCP server implementations. **When NOT to choose:** Don't choose emcee if you need to expose APIs requiring authentication for specification access, or if you need advanced features like resources, prompts, and sampling which aren't yet supported. ### [gopher-mcp](https://mcpcatalogs.com/en/server/gophersecurity-gopher-mcp) by GopherSecurity · score 47 · ★114 A comprehensive C++ SDK for MCP with enterprise features and multi-language bindings. **When to choose:** Choose gopher-mcp when you need maximum performance, low latency, or require integration into C++ environments where Python/Node.js is not feasible. **When NOT to choose:** Avoid gopher-mcp if you're developing for platforms without C++ support, or if you prefer higher-level languages with simpler development workflows and don't need the performance benefits of C++. ### [context-awesome](https://mcpcatalogs.com/en/server/bh-rat-context-awesome) by bh-rat · score 47 · ★51 An MCP server providing access to curated awesome lists with search and section discovery tools. **When to choose:** Choose Context Awesome when your AI agents need access to curated, high-quality technical resources across thousands of topics without web search. **When NOT to choose:** Don't choose this for highly specialized niche topics not covered in awesome lists, or if you need real-time, dynamic content beyond what's captured in static lists. ### [deploystack](https://mcpcatalogs.com/en/server/deploystackio-deploystack) by deploystackio · score 47 · ★57 DeployStack transforms stdio-only MCP servers into HTTP endpoints for automation platforms. **When to choose:** Choose DeployStack when you need to host MCP servers as HTTP endpoints for integration with workflow tools like n8n, Dify, or Voiceflow, or when you need team management features with RBAC and credential management for MCP servers. **When NOT to choose:** Avoid DeployStack if you only need simple stdio-based MCP servers without HTTP requirements, if you need full control over deployment environment, or if you prefer open-source without any proprietary components. ### [wasmcp](https://mcpcatalogs.com/en/server/wasmcp-wasmcp) by wasmcp · score 46 · ★79 A WebAssembly Component SDK for building composable MCP servers with Rust, Python and TypeScript support. **When to choose:** Choose wasmCP when you need highly modular, sandboxed MCP servers with isolated components, especially in security-sensitive or multi-tenant environments. **When NOT to choose:** Don't choose wasmCP if you need a traditional monolithic server, if WebAssembly overhead is unacceptable for your use case, or if you're in an environment without WASM runtime support. ### [deterministic-agent-control-protocol](https://mcpcatalogs.com/en/server/elliot35-deterministic-agent-control-protocol) by elliot35 · score 46 · ★86 A governance gateway for AI agents providing bounded, auditable control via MCP proxy, shell proxy, and HTTP API. **When to choose:** When you need to govern AI agents with strict access controls, audit trails, and policy enforcement for sensitive operations. **When NOT to choose:** If you're working with open-source LLMs not connected to MCP clients, or if you need minimal governance overhead. ### [zeromcp](https://mcpcatalogs.com/en/server/mrexodia-zeromcp) by mrexodia · score 46 · ★77 Zero-dependency Python MCP server implementation with type-safe tool definitions and multiple transport options. **When to choose:** Choose zeromcp when you need a lightweight, dependency-free MCP server for simple tools, especially when working in restricted environments where minimizing dependencies is crucial. **When NOT to choose:** Avoid zeromcp if you need advanced MCP features, complex tool definitions, or require authentication mechanisms as it's intentionally minimal and lacks these capabilities. ### [sugar](https://mcpcatalogs.com/en/server/roboticforce-sugar) by roboticforce · score 46 · ★75 Sugar provides persistent memory for AI coding agents with MCP integration for cross-session context and task automation. **When to choose:** Choose Sugar when you need persistent memory across AI coding sessions and want to automate GitHub issue resolution while maintaining coding consistency across projects. **When NOT to choose:** Don't choose Sugar if you need a commercial license for proprietary use, as the AGPL-3.0 license requires source code disclosure for modifications. ### [ai-command](https://mcpcatalogs.com/en/server/mcp-wp-ai-command) by mcp-wp · score 46 · ★116 WordPress MCP server enabling AI-powered site management through WP-CLI integration. **When to choose:** Choose this if you're already working with WordPress and want to integrate AI capabilities directly through WP-CLI during development. **When NOT to choose:** Avoid if you don't use WordPress or if you need MCP server functionality outside of WordPress environments. ### [ToolsForMCPServer](https://mcpcatalogs.com/en/server/tanaikech-toolsformcpserver) by tanaikech · score 46 · ★105 An MCP server built with Google Apps Script providing 160+ tools for Google Workspace integration. **When to choose:** Choose this if you're already using Google Workspace and want to extend AI capabilities with custom tools without building complex authentication systems. **When NOT to choose:** Avoid if you need tools outside of Google Workspace ecosystem, or if you prefer a self-hosted solution without dependency on Google's infrastructure. ### [Unity-AI-Animation](https://mcpcatalogs.com/en/server/ivanmurzak-unity-ai-animation) by IvanMurzak · score 46 · ★78 Unity AI Animation MCP server enables AI-powered animation workflows through natural language commands for AnimationClips and AnimatorControllers. **When to choose:** Unity developers who want to automate animation workflows and create complex state machines through natural language commands. **When NOT to choose:** Non-Unity projects or developers who need advanced animation features not covered by this MCP server. ### [SageFs](https://mcpcatalogs.com/en/server/willehrendreich-sagefs) by WillEhrendreich · score 46 · ★64 A live F# development engine with hot reload, live testing, and MCP integration for AI agents. **When to choose:** Choose SageFs when working with F# projects, especially web development with frameworks like Falco, and need immediate feedback through live testing and hot reload with AI agent integration. **When NOT to choose:** Not suitable if you primarily work with languages other than F#, or if you need type redefinition with browser hot reload (a limitation of .NET runtime). ### [total-agent-memory](https://mcpcatalogs.com/en/server/vbcherepanov-total-agent-memory) by vbcherepanov · score 46 · ★34 Persistent memory layer for AI coding agents with knowledge graphs, embeddings, and 3D visualization. **When to choose:** Choose total-agent-memory when you need persistent memory for AI coding agents with strong local privacy guarantees and want to preserve context across sessions without cloud dependencies. **When NOT to choose:** Don't choose this if you need cloud-based memory synchronization across multiple machines or require a fully managed SaaS solution without any self-hosting requirements. ### [ai-software-architect](https://mcpcatalogs.com/en/server/codenamev-ai-software-architect) by codenamev · score 46 · ★65 AI-powered architecture documentation framework with ADRs and MCP server integration for Claude and Cursor. **When to choose:** Choose this if you're using Claude Code or Cursor and want programmatic architecture documentation with AI integration. **When NOT to choose:** Avoid if you need offline functionality or full recalibration features, as these are only available in the traditional setup. ### [Gearlynx](https://mcpcatalogs.com/en/server/drhelius-gearlynx) by drhelius · score 46 · ★65 Gearlynx is a cross-platform Atari Lynx emulator with embedded MCP server for AI-assisted debugging. **When to choose:** Choose Gearlynx when working with Atari Lynx game development, reverse engineering, or when you need AI assistance debugging vintage console software with precise hardware emulation. **When NOT to choose:** Don't choose Gearlynx if you need emulation of gaming systems other than Atari Lynx, or if you require a more general-purpose debugging framework with broader platform support. ### [estonia-ai-kit](https://mcpcatalogs.com/en/server/stefanoamorelli-estonia-ai-kit) by stefanoamorelli · score 46 · ★65 Estonia AI Kit provides MCP servers and skills to connect AI models to Estonia's digital infrastructure and government services. **When to choose:** Choose this if you need to connect AI models to Estonia's digital infrastructure, especially if you're working with government services like business registry, tax, or open data. **When NOT to choose:** Don't choose this if you need services outside Estonia, as this toolkit is specifically designed for the Estonian digital ecosystem only. ### [openclaw-superpowers](https://mcpcatalogs.com/en/server/archieindian-openclaw-superpowers) by ArchieIndian · score 46 · ★62 A skill library for OpenClaw with 56 plug-and-play skills including cron scheduling, security guardrails, and MCP health monitoring. **When to choose:** Choose this if you're building persistent autonomous AI agents with OpenClaw and need comprehensive skills for 24/7 operation, security, and self-improvement capabilities. **When NOT to choose:** Don't choose this if you're not using OpenClaw as your agent runtime, as these skills are specifically designed for OpenClaw's persistent architecture and won't work with other frameworks. ### [mcp-shell](https://mcpcatalogs.com/en/server/sonirico-mcp-shell) by sonirico · score 46 · ★77 MCP server for secure and auditable shell command execution with configurable security policies. **When to choose:** When you need AI agents to safely interact with the local filesystem and system commands with precise control over execution parameters and security boundaries. **When NOT to choose:** If you require complex shell scripting features or need to run commands with full system privileges without security restrictions. ### [odoo-mcp-pro](https://mcpcatalogs.com/en/server/pantalytics-odoo-mcp-pro) by pantalytics · score 46 · ★38 MCP server connecting AI tools to Odoo ERP for natural language data queries and management. **When to choose:** Choose this when you need natural language access to Odoo business data and already use Odoo as your ERP system. **When NOT to choose:** Avoid if you need multi-tenant support or don't want to comply with Elastic License 2.0 restrictions on offering hosted services to third parties. ### [model-context-protocol-resources](https://mcpcatalogs.com/en/server/cyanheads-model-context-protocol-resources) by cyanheads · score 46 · ★279 A comprehensive collection of MCP guides, utilities, and server implementations for practical exploration of the Model Context Protocol. **When to choose:** Choose this resource when you're learning MCP development or need comprehensive examples of different server implementations across various domains. **When NOT to choose:** Not suitable if you need a single, production-ready MCP server with extensive documentation and support, as this is a collection of resources rather than a specific implementation. ### [adbfriend](https://mcpcatalogs.com/en/server/mikepenz-adbfriend) by mikepenz · score 46 · ★71 Android ADB CLI tool with integrated MCP Server for common development actions. **When to choose:** Choose this when you need programmatic control over Android devices during development or testing, especially if you're already using Claude Desktop or other MCP-compatible AI tools. **When NOT to choose:** Not suitable for non-developer Android management tasks or when you need fine-grained permission controls beyond what ADB provides. ### [deno-mcp-template](https://mcpcatalogs.com/en/server/phughesmcr-deno-mcp-template) by phughesmcr · score 46 · ★30 Comprehensive TypeScript template for building production MCP servers with Deno, supporting multiple distribution formats. **When to choose:** Choose this template if you want to build production MCP servers with TypeScript and Deno, especially when you need multiple distribution formats and built-in infrastructure like persistent state and sandboxed execution. **When NOT to choose:** Don't choose this template if you're not using TypeScript/Deno, or if you need more specialized functionality beyond what the template provides (like custom authentication systems beyond bearer tokens). ### [teamcity-mcp](https://mcpcatalogs.com/en/server/daghis-teamcity-mcp) by Daghis · score 46 · ★25 MCP server for JetBrains TeamCity that provides 87 tools to control builds, tests, agents and configurations from AI coding assistants. **When to choose:** Choose this server if you need comprehensive management of TeamCity configurations beyond basic build triggering, or if you're using a TeamCity version older than 2026.1. **When NOT to choose:** Don't choose this if you only need basic build triggering and log reading in TeamCity 2026.1+, as the built-in MCP endpoint is sufficient and requires no separate installation. ### [registry](https://mcpcatalogs.com/en/server/mcparmory-registry) by mcparmory · score 46 · ★25 Production-ready MCP servers for 70+ APIs generated from OpenAPI specs with comprehensive tool coverage. **When to choose:** Choose MCP Armory when you need quickly deployable, production-ready MCP servers for popular APIs without custom development, especially when you want standardized implementations with built-in authentication and validation. **When NOT to choose:** Don't choose this if you need custom API endpoints beyond what's available in the registry, or if you require specialized authentication methods not covered by the generated servers. ### [editor-mcp-server](https://mcpcatalogs.com/en/server/playcanvas-editor-mcp-server) by playcanvas · score 46 · ★108 MCP server for automating PlayCanvas editor through entity and asset management tools. **When to choose:** Choose this server if you're working with PlayCanvas and want to automate scene setup, asset management, or entity manipulation through AI agents. **When NOT to choose:** Avoid if you need to work with other 3D engines or require multi-editor support, as this is specifically designed for PlayCanvas only. ### [mcp-server-atlassian-jira](https://mcpcatalogs.com/en/server/aashari-mcp-server-atlassian-jira) by aashari · score 46 · ★69 TypeScript MCP server enabling AI systems to interact with Atlassian Jira via REST API for project management and issue tracking. **When to choose:** Choose this MCP server if your team uses Atlassian Jira and wants AI assistants to directly interact with project management and issue tracking workflows. **When NOT to choose:** Don't choose this if you use Jira Server/Data Center (on-premise) instead of Jira Cloud, or need more specialized Jira operations beyond the generic API tools provided. ### [snippy](https://mcpcatalogs.com/en/server/azure-samples-snippy) by Azure-Samples · score 46 · ★110 Azure Functions-based MCP server with vector search and multi-agent workflows for AI-powered code snippet management. **When to choose:** Choose Snippy if you're building MCP tools on Azure and need a reference implementation that demonstrates multi-agent orchestration and vector search for code-related applications. **When NOT to choose:** Avoid Snippy if you need a vendor-neutral solution, are not using Azure services, or require extensive customization beyond the provided multi-agent workflow pattern. ### [sub-agents-mcp](https://mcpcatalogs.com/en/server/shinpr-sub-agents-mcp) by shinpr · score 46 · ★83 Define task-specific AI sub-agents in Markdown for use with any MCP-compatible tool. **When to choose:** Choose this when you want to reuse Claude Code-style sub-agents across different AI tools or share specialized agent definitions with team members regardless of their preferred IDE. **When NOT to choose:** Avoid if you need only simple agent definitions or if you prefer a more integrated solution without requiring separate MCP configuration. ### [oh-my-codex](https://mcpcatalogs.com/en/server/scalarian-oh-my-codex) by scalarian · score 46 · ★61 Orchestration layer for OpenAI Codex with MCP integration for task, state, memory and team management. **When to choose:** Choose oh-my-codex if you're using OpenAI Codex for development and need persistent state, team coordination, and structured workflows for tasks like code review, planning, and implementation. **When NOT to choose:** Don't choose oh-my-codex if you need support for AI models other than Codex, or if you're looking for a general-purpose AI assistant without the specific development workflow orchestration it provides. ### [mcp-batchit](https://mcpcatalogs.com/en/server/ryanjoachim-mcp-batchit) by ryanjoachim · score 46 · ★59 MCP BatchIt aggregates multiple tool calls into single batch requests, reducing overhead and token usage in AI agent workflows. **When to choose:** When working with AI agents that need to make multiple tool calls and you want to minimize token usage and network overhead. **When NOT to choose:** If you have complex workflows where one tool's output depends on another's result, as BatchIt doesn't support data passing between operations in the same batch. ### [mymir](https://mcpcatalogs.com/en/server/frkak-mymir) by FrkAk · score 46 · ★68 Mymir is an AI-native project management tool that provides MCP tools for task management and context tracking in agentic engineering workflows. **When to choose:** Choose Mymir when working with AI coding assistants and need persistent project management with task dependencies and context retrieval, especially if you're already using Claude Code, Codex, Cursor, or Gemini. **When NOT to choose:** Avoid Mymir if you need a standalone project management tool without AI integration, or if you prefer a self-hosted solution without PostgreSQL dependency. ### [julia-mcp](https://mcpcatalogs.com/en/server/aplavin-julia-mcp) by aplavin · score 46 · ★63 MCP server for persistent Julia sessions, avoiding startup costs and preserving state between calls. **When to choose:** Choose julia-mcp when you need efficient Julia code execution with state persistence across multiple interactions, especially for Julia package development or scientific computing workflows. **When NOT to choose:** Avoid if you need Julia script execution with completely isolated runs or if you're working in an environment where Python is not available (the MCP server itself is written in Python). ### [plan-cascade](https://mcpcatalogs.com/en/server/taoidle-plan-cascade) by Taoidle · score 46 · ★80 Plan Cascade MCP server enables cascading development workflows by breaking down complex projects into parallel executable tasks with AI agents. **When to choose:** Choose Plan Cascade when working on complex projects that benefit from decomposition into parallel tasks, especially if you're already using Claude Code or need visual workflow management. **When NOT to choose:** Avoid if you need a simple coding assistant for small tasks, as Plan Cascade is designed for complex project orchestration with significant setup overhead. ### [indxr](https://mcpcatalogs.com/en/server/bahdotsh-indxr) by bahdotsh · score 46 · ★64 A fast codebase indexer and knowledge wiki for AI agents with 26 MCP tools for code analysis. **When to choose:** Choose indxr when you need comprehensive code understanding tools with persistent knowledge capture, especially for larger codebases where maintaining up-to-date documentation is challenging. **When NOT to choose:** Don't choose indxr if you need write access to code (it's read-only), have very simple codebases that don't warrant complex indexing, or prefer a more integrated IDE solution. ### [mcp4k](https://mcpcatalogs.com/en/server/ondrsh-mcp4k) by ondrsh · score 46 · ★69 Kotlin Multiplatform framework for building MCP clients and servers with compile-time type safety. **When to choose:** Choose mcp4k when building MCP servers/clients in Kotlin Multiplatform projects and you need compile-time type safety and automatic JSON-RPC handling. **When NOT to choose:** Not for projects requiring transports beyond stdio, or when you need to use a language other than Kotlin. ### [Lambda-MCP-Server](https://mcpcatalogs.com/en/server/mikegc-aws-lambda-mcp-server) by mikegc-aws · score 46 · ★232 AWS Lambda serverless MCP server implementation with session management and HTTP streaming support. **When to choose:** Choose this when you need to deploy MCP tools serverlessly on AWS and want minimal boilerplate with good developer experience. **When NOT to choose:** Avoid if you're not using AWS ecosystem, as this is tightly coupled to AWS Lambda and related services. ### [mcp-debug](https://mcpcatalogs.com/en/server/giantswarm-mcp-debug) by giantswarm · score 46 · ★37 A command-line tool for debugging MCP servers with OAuth 2.1 authentication, interactive REPL, and server mode. **When to choose:** Choose mcp-debug when you need a dedicated debugging tool for MCP servers with comprehensive OAuth 2.1 support and detailed JSON-RPC message logging. **When NOT to choose:** Don't choose mcp-debug if you need support for non-HTTP transports, as it only works with `streamable-http` protocol MCP servers. ### [bw-modeling-mcp](https://mcpcatalogs.com/en/server/dnic-dev-bw-modeling-mcp) by dnic-dev · score 46 · ★33 MCP server for AI-assisted development in SAP BW/4HANA systems with comprehensive modeling tools. **When to choose:** Choose this if you work with SAP BW/4HANA systems and want AI assistants to directly interact with your BW environment for development, analysis, or data lineage tasks. **When NOT to choose:** Don't choose this if you're using BW 7.5 (on HANA) instead of BW/4HANA, or if you need functionality beyond BW modeling objects and data operations. ### [stacklit](https://mcpcatalogs.com/en/server/glincker-stacklit) by glincker · score 46 · ★81 Stacklit provides an MCP server that scans codebases to generate compact indexes for AI coding assistants. **When to choose:** Choose Stacklit when working with AI coding assistants on large codebases where context window efficiency is critical and you need structured code navigation without overwhelming the agent with raw code. **When NOT to choose:** Don't choose Stacklit if you're working with very small codebases where the overhead of maintaining an index outweighs the benefits, or if you need real-time code analysis without any pre-processing. ### [mcp-gateway](https://mcpcatalogs.com/en/server/mikkoparkkola-mcp-gateway) by MikkoParkkola · score 46 · ★25 Universal MCP Gateway that multiplexes tools with 95% token savings via Meta-MCP. **When to choose:** Choose MCP Gateway when you have multiple MCP servers and want to reduce context overhead, or when you need to connect REST APIs without building custom MCP servers. **When NOT to choose:** Don't choose MCP Gateway if you only have a small number of tools and servers, or if you need direct, low-latency connections to individual backend services without the gateway layer. ### [roundtable](https://mcpcatalogs.com/en/server/askbudi-roundtable) by askbudi · score 46 · ★107 Zero-configuration MCP server coordinating multiple AI coding assistants through intelligent auto-discovery. **When to choose:** When you need to leverage multiple specialized AI models simultaneously for complex engineering tasks without manually switching between tools. **When NOT to choose:** If you're working offline or in environments with limited internet connectivity, as Roundtable requires access to various AI model APIs. ### [lazy-mcp](https://mcpcatalogs.com/en/server/voicetreelab-lazy-mcp) by voicetreelab · score 46 · ★87 Lazy MCP proxy server that loads tools on-demand to reduce context window usage and token consumption. **When to choose:** Choose Lazy MCP when you need many MCP tools but face context window limitations, or when you require fine-grained permission control for sensitive operations across multiple MCP servers. **When NOT to choose:** Don't choose Lazy MCP if you have minimal MCP tools that fit within your context window, or if you can't afford the added complexity of a proxy layer. ### [codeTree](https://mcpcatalogs.com/en/server/thinkyminer-codetree) by ThinkyMiner · score 46 · ★25 MCP server with 23 tools for structured code understanding via tree-sitter across 10 languages. **When to choose:** Choose codeTree when you need efficient code analysis for AI agents working across multiple languages without the overhead of vector databases or complex configuration. **When NOT to choose:** Don't choose codeTree if you need real-time code modification capabilities or are working with languages not in its 10-language support list. ### [schedcp](https://mcpcatalogs.com/en/server/eunomia-bpf-schedcp) by eunomia-bpf · score 46 · ★105 SchedCP enables AI optimization of Linux kernel schedulers using MCP Server for automatic management and performance tuning. **When to choose:** Choose SchedCP if you need AI-powered optimization for Linux scheduler performance and have experimental kernel access. **When NOT to choose:** Avoid this if you need a production-ready solution or cannot use Linux kernels 6.12+ with sched-ext support. ### [claude-code-emacs](https://mcpcatalogs.com/en/server/yuya373-claude-code-emacs) by yuya373 · score 46 · ★63 Emacs package providing MCP integration for Claude Code, enabling AI-powered coding sessions with buffer and LSP operations. **When to choose:** Choose this if you're an Emacs user who wants to integrate Claude Code directly into your editing environment without switching contexts. **When NOT to choose:** Not ideal if you don't use Emacs or prefer a more generalized IDE with AI capabilities, as this solution is tightly coupled with Emacs. ### [nautex](https://mcpcatalogs.com/en/server/hmldns-nautex) by hmldns · score 46 · ★79 MCP server for guiding coding agents through requirements to implementation using Nautex AI. **When to choose:** Choose this if you're already using or planning to use Nautex AI for requirements gathering and want structured implementation guidance for your coding agents. **When NOT to choose:** Avoid if you prefer open-source solutions without vendor lock-in, or if you're not comfortable with cloud-based requirements management. ### [Vera](https://mcpcatalogs.com/en/server/lemon07r-vera) by lemon07r · score 46 · ★82 Local code search tool combining BM25, vector similarity, and cross-encoder reranking with MCP support. **When to choose:** Choose Vera when you need powerful, local code search with excellent accuracy for navigating large codebases and understanding code structure. **When NOT to choose:** Don't choose Vera if you need real-time collaborative search capabilities or require searching codebases hosted on remote repositories without local indexing. ### [mcpshim](https://mcpcatalogs.com/en/server/mcpshim-mcpshim) by mcpshim · score 46 · ★59 MCPShim turns remote MCP servers into local CLI commands, handling auth and tool invocation in one daemon. **When to choose:** Choose MCPShim when you need to integrate multiple MCP servers into shell workflows or AI agents without implementing custom authentication and session management for each one. **When NOT to choose:** Don't choose MCPShim if you need direct programmatic access to MCP servers without the CLI abstraction layer, or if you're building a tightly-coupled application where the overhead of the daemon isn't justified. ### [wmux](https://mcpcatalogs.com/en/server/openwong2kim-wmux) by openwong2kim · score 46 · ★76 Windows terminal multiplexer with MCP server integration for AI coding agents like Claude, supporting split terminals and browser automation. **When to choose:** Choose wmux if you need to run multiple AI coding agents simultaneously on Windows with browser automation capabilities and want session persistence without WSL. **When NOT to choose:** Don't choose wmux if you're on macOS or Linux (where tmux is native), or if you need terminal multiplexing without AI agent integration. ### [vscode-mcp](https://mcpcatalogs.com/en/server/tjx666-vscode-mcp) by tjx666 · score 46 · ★76 VSCode MCP server that provides real-time LSP diagnostics, type information, and code navigation for AI coding agents without waiting for slow tsc/eslint checks. **When to choose:** Choose this when you're using AI coding assistants (like Claude Code or Cursor) and need fast access to VSCode's LSP information without waiting for slow build tools. **When NOT to choose:** Don't choose this if you need to use non-VSCode editors or if you're concerned about security risks from the execute_command tool that can run arbitrary VSCode commands. ### [mcp.zig](https://mcpcatalogs.com/en/server/muhammad-fiaz-mcp-zig) by muhammad-fiaz · score 46 · ★35 A comprehensive MCP library for Zig that supports both server and client implementations with protocol compliance. **When to choose:** Choose mcp.zig when you need high-performance MCP implementations in Zig, especially for server development where Zig's performance and safety features are desired. **When NOT to choose:** Don't choose mcp.zig if you're not working with Zig or need extensive ecosystem libraries that only exist for Python/TypeScript MCP implementations. ### [laravel-loop](https://mcpcatalogs.com/en/server/kirschbaum-development-laravel-loop) by kirschbaum-development · score 46 · ★128 Laravel Loop is an MCP server connecting Laravel applications with AI assistants via tools for models, factories, and Stripe integration. **When to choose:** Choose Laravel Loop if you're building Laravel applications and want to integrate AI assistants with your application data and functionality through the MCP protocol. **When NOT to choose:** Avoid if you're not using Laravel or need a stable production-ready solution, as Laravel Loop is still in beta with limited write capabilities. ### [Sephera](https://mcpcatalogs.com/en/server/reim-developer-sephera) by Reim-developer · score 46 · ★77 Fast Rust CLI for code metrics and AST-compressed context bundles with built-in MCP server. **When to choose:** Choose Sephera when you need fast codebase analysis with AST compression for LLM contexts and want to expose these capabilities directly to AI agents through MCP without complex wrappers. **When NOT to choose:** Avoid Sephera if you need write access to databases (it's read-only only), or if you require a more general-purpose agent runtime rather than a specialized code analysis tool. ### [mcp-server-circleci](https://mcpcatalogs.com/en/server/circleci-public-mcp-server-circleci) by CircleCI-Public · score 46 · ★83 A TypeScript MCP server that integrates CircleCI's development workflow with AI assistants via natural language commands. **When to choose:** Choose this server if you're using CircleCI for CI/CD and want to interact with your pipelines through natural language commands in your IDE. **When NOT to choose:** Avoid if you're not using CircleCI or prefer a more generic CI/CD tool with broader platform support. ### [mcp-n8n-builder](https://mcpcatalogs.com/en/server/spences10-mcp-n8n-builder) by spences10 · score 46 · ★83 MCP server for programmatic creation and management of n8n workflows with API tools. **When to choose:** Choose this MCP server if you're already using n8n and want to enable AI assistants to programmatically manage workflows without direct human intervention. **When NOT to choose:** Don't choose this if you're not using n8n as your workflow automation platform, as it's tightly coupled to n8n's API and workflow structure. ### [mcpls](https://mcpcatalogs.com/en/server/bug-ops-mcpls) by bug-ops · score 46 · ★34 Universal MCP to LSP bridge that exposes Language Server Protocol capabilities as MCP tools for AI agents. **When to choose:** Choose mcpls when you need AI coding assistants to have deep, structured understanding of your codebase beyond text processing, with access to type systems, cross-references, and real diagnostics. **When NOT to choose:** Don't choose mcpls if you're working primarily with non-LSP supported languages or need write capabilities beyond refactoring, as it focuses on exposing read-only LSP intelligence through MCP. ### [ls-mcp](https://mcpcatalogs.com/en/server/lirantal-ls-mcp) by lirantal · score 46 · ★83 CLI tool to discover and analyze MCP server configurations across various AI applications and IDEs. **When to choose:** Choose ls-mcp when you need to audit, debug, or manage MCP server configurations across multiple AI applications and development environments. **When NOT to choose:** Don't choose ls-mcp if you need to actually interact with MCP servers rather than just listing their configurations. ### [ai-forge-mcp](https://mcpcatalogs.com/en/server/hurtzdonutstudios-ai-forge-mcp) by HurtzDonutStudios · score 46 · ★59 Comprehensive MCP package with 565 tools controlling AAA game asset pipeline across 16 servers. **When to choose:** Developers working with Unreal Engine 5 and Blender who want to automate entire game asset production pipelines and require professional-grade tools for complex 3D workflows. **When NOT to choose:** Individuals looking for open-source solutions or on-premise installations, as AI Forge appears to be a commercial product with vendor lock-in potential. ### [funplay-unity-mcp](https://mcpcatalogs.com/en/server/funplayai-funplay-unity-mcp) by FunplayAI · score 46 · ★57 Advanced MCP server for Unity Editor with 91 built-in tools for automation, scene creation, and AI development. **When to choose:** Choose this when working with Unity projects and need AI agents to directly interact with the editor through comprehensive tooling for scene creation, script generation, and automation. **When NOT to choose:** Avoid if you need runtime game functionality (this is Editor-only) or if you prefer a Python-based MCP server architecture rather than a Unity package. ### [Uni-CLI](https://mcpcatalogs.com/en/server/olo-dot-io-uni-cli) by olo-dot-io · score 46 · ★82 Universal CLI substrate providing 235+ websites, desktop apps, and MCP servers as searchable commands for AI agents. **When to choose:** Choose Uni-CLI when your AI agent needs broad access to web services, desktop applications, and external CLIs through a unified interface with repair capabilities. **When NOT to choose:** Avoid Uni-CLI if you need write access to databases or systems, require specialized domain-specific tools not in its catalog, or need a lightweight solution with minimal dependencies. ### [trace-mcp](https://mcpcatalogs.com/en/server/nikolai-vysotskyi-trace-mcp) by nikolai-vysotskyi · score 46 · ★71 MCP server that eliminates AI agent recomputation by building and serving cross-language dependency graphs through Model Context Protocol. **When to choose:** Choose trace-mcp when working with large, multi-language codebases where you need to understand dependencies across frameworks and reduce redundant agent processing. **When NOT to choose:** Don't choose trace-mcp for small projects or when you need real-time code analysis without the initial indexing overhead. ### [sift-gateway](https://mcpcatalogs.com/en/server/lourencomaciel-sift-gateway) by lourencomaciel · score 46 · ★29 Sift is a reliability gateway for MCP and CLI tools that stores artifacts and returns schema-stable JSON outputs. **When to choose:** Choose Sift when working with MCP tools that produce large outputs, handle sensitive data, or require pagination completion, especially when you need to maintain data reliability while minimizing context usage. **When NOT to choose:** Avoid Sift if you need low-latency tool responses, have very simple API interactions without pagination concerns, or prefer direct tool access without the additional abstraction layer. ### [typo3-mcp-server](https://mcpcatalogs.com/en/server/hauptsachenet-typo3-mcp-server) by hauptsacheNet · score 46 · ★80 MCP server for TYPO3 CMS enabling safe AI content management through workspaces with OAuth authentication. **When to choose:** Choose this if you're using TYPO3 CMS and want AI assistance for content management, SEO optimization, and multilingual site management with safe workspace-based operations. **When NOT to choose:** Avoid if you need fileadmin support or workspace selection features, as these are currently missing, or if you're not using TYPO3 CMS. ### [mcp-stdio-guard](https://mcpcatalogs.com/en/server/1utkarsh1-mcp-stdio-guard) by 1Utkarsh1 · score 46 · ★57 A tool that validates MCP stdio servers by catching stdout pollution and handshake failures before clients encounter them. **When to choose:** You should choose this tool when developing or maintaining MCP stdio servers to prevent common protocol failures caused by stdout pollution and handshake issues. **When NOT to choose:** Don't use this for servers using non-stdio transports (HTTP/WebSocket), or if you need runtime validation beyond the initial handshake phase. ### [nutrient-dws-mcp-server](https://mcpcatalogs.com/en/server/pspdfkit-nutrient-dws-mcp-server) by PSPDFKit · score 46 · ★63 MCP server connecting AI agents to PDF processing capabilities via Nutrient DWS API for document manipulation. **When to choose:** Choose this server when you need AI agents to manipulate documents beyond just reading them, especially if you require document conversion, signing, redaction, or OCR capabilities. **When NOT to choose:** Avoid this if you're working with sensitive documents that can't leave your local environment, as it uploads documents to a third-party service for processing. ### [hyperterse](https://mcpcatalogs.com/en/server/hyperterse-hyperterse) by hyperterse · score 46 · ★82 Agentic server framework that combines MCP tools, prompts, resources, and agents in a single deployable service with Go. **When to choose:** Choose Hyperterse when you need to build a comprehensive service that combines MCP tools, prompts, resources, and agents in a single deployable unit, especially when working with existing databases and requiring production features like authentication and monitoring. **When NOT to choose:** Don't choose Hyperterse if you only need basic MCP tools without agents, or if you prefer a simpler framework with less boilerplate and more focused functionality. ### [mcp-proxy-server](https://mcpcatalogs.com/en/server/ptbsare-mcp-proxy-server) by ptbsare · score 46 · ★194 A central proxy server that aggregates and manages multiple MCP servers through a unified HTTP or SSE endpoint with web UI. **When to choose:** Choose this when you need to manage multiple MCP servers centrally and want a web interface for configuration and monitoring. **When NOT to choose:** Avoid if you need a simple, single MCP server setup or if you're concerned about the security implications of the web terminal feature. ### [rust-mcp-schema](https://mcpcatalogs.com/en/server/rust-mcp-stack-rust-mcp-schema) by rust-mcp-stack · score 46 · ★75 Type-safe Rust implementation of MCP protocol schemas, automatically generated from official specifications. **When to choose:** Choose this when building MCP servers or clients in Rust requiring type safety and compatibility with multiple MCP protocol versions. **When NOT to choose:** Not suitable if you need a high-level framework for building MCP servers - look at rust-mcp-sdk instead. ### [mcp-design-system-extractor](https://mcpcatalogs.com/en/server/freema-mcp-design-system-extractor) by freema · score 46 · ★64 MCP server that extracts HTML, styles and component metadata from Storybook design systems to help with development. **When to choose:** Choose this MCP server when working with Storybook-based design systems and need AI assistance to analyze, extract, or reproduce component implementations. **When NOT to choose:** Don't use this if your Storybook instance requires authentication beyond basic URL access or if you need to extract information from non-Storybook design systems. ### [mcp-openapi-schema-explorer](https://mcpcatalogs.com/en/server/kadykov-mcp-openapi-schema-explorer) by kadykov · score 46 · ★71 An MCP server for token-efficient exploration of OpenAPI/Swagger specs via MCP Resource Templates. **When to choose:** Choose this when you need efficient token-based access to OpenAPI/Swagger specs in AI clients without loading entire specification files into context. **When NOT to choose:** Don't choose this if you need write access to API specifications or require direct API endpoint interaction rather than documentation exploration. ### [Kaimon.jl](https://mcpcatalogs.com/en/server/kahliburke-kaimon-jl) by kahliburke · score 46 · ★67 MCP server providing AI agents full access to Julia's runtime with code execution, debugging, testing, and semantic search. **When to choose:** Choose Kaimon when working with Julia codebases and needing comprehensive AI access to runtime execution, debugging, and semantic analysis capabilities. **When NOT to choose:** Avoid Kaimon if you're not using Julia, as it only provides integration with the Julia ecosystem, not with other programming languages. ### [csharp-runner](https://mcpcatalogs.com/en/server/sdcb-csharp-runner) by sdcb · score 46 · ★109 A fast, secure C# code execution platform with container isolation and MCP protocol support. **When to choose:** Choose C# Runner when you need a secure, isolated environment for executing C# code, especially if you're already using Docker and want both HTTP API and MCP protocol support. **When NOT to choose:** Don't choose this if you need to execute code in languages other than C#, or if you require more granular security controls beyond container isolation. ### [ipybox](https://mcpcatalogs.com/en/server/gradion-ai-ipybox) by gradion-ai · score 46 · ★74 Unified execution environment for Python, shell commands, and MCP tool calls with IPython kernels. **When to choose:** Choose ipybox when you need a unified environment for Python, shell commands, and MCP tools with sandboxing and approval controls, especially for building secure AI applications. **When NOT to choose:** Avoid ipybox if you need pure JavaScript/TypeScript execution environments or if you're looking for a managed cloud service rather than a local execution solution. ### [bridge4simulator](https://mcpcatalogs.com/en/server/appgram-bridge4simulator) by AppGram · score 45 · ★81 MCP server controlling iOS Simulator with extensive automation tools for testing and development. **When to choose:** Choose this server when you need AI-powered iOS app testing, automation, or UI interaction capabilities and are already working within the Apple ecosystem. **When NOT to choose:** Not ideal for non-Apple environments or if you need device testing on physical iOS devices rather than simulators. ### [pydantic-rpc](https://mcpcatalogs.com/en/server/i2y-pydantic-rpc) by i2y · score 45 · ★74 PydanticRPC rapidly exposes Pydantic models as gRPC, ConnectRPC, and MCP services without protobuf files. **When to choose:** Choose PydanticRPC when you need to expose Python services with strong typing via multiple protocols and want to avoid manual protobuf definitions. **When NOT to choose:** Avoid if you need fine-grained control over generated protobuf files or require advanced features like comprehensive observability built-in. ### [github-to-mcp](https://mcpcatalogs.com/en/server/nirholas-github-to-mcp) by nirholas · score 45 · ★28 Converts GitHub repositories into MCP servers by extracting tools from OpenAPI, GraphQL & REST APIs. **When to choose:** Choose this tool when you need to quickly create MCP servers from existing code repositories, especially when working with OpenAPI, GraphQL, or REST APIs, and want to minimize manual tool definition. **When NOT to choose:** Avoid this if you need highly customized tools with complex business logic beyond what can be inferred from repository metadata and API specifications. ### [MCPApp](https://mcpcatalogs.com/en/server/tanaikech-mcpapp) by tanaikech · score 45 · ★47 Google Apps Script-based MCP server implementation for AI clients with tools and prompts support. **When to choose:** Choose MCPApp when you need to implement MCP functionality within Google's ecosystem or want a lightweight serverless solution for AI integration. **When NOT to choose:** Avoid MCPApp if you need heavy processing capabilities, custom authentication systems, or want to avoid Google's ecosystem restrictions. ### [csv-editor](https://mcpcatalogs.com/en/server/santoshray02-csv-editor) by santoshray02 · score 45 · ★23 Stateful CSV editing MCP server with 39 pandas-powered tools, sessions, undo/redo and auto-save features. **When to choose:** You need an AI assistant to edit and transform CSV files with stateful operations, undo history, and auto-save capabilities, not just analyze them. **When NOT to choose:** For read-only analytics on multi-GB files, DuckDB/Polars-based MCPs are more suitable; CSV Editor's pandas implementation is optimized for smaller datasets. ### [roampal-core](https://mcpcatalogs.com/en/server/roampal-ai-roampal-core) by roampal-ai · score 45 · ★46 Outcome-based persistent memory MCP server that improves AI assistants through conversational learning. **When to choose:** Choose Roampal when using Claude Code or OpenCode and want to create a persistent memory system that learns from successful coding patterns and user preferences. **When NOT to choose:** Not ideal if you need integration with other AI coding assistants beyond Claude Code and OpenCode, or if you prefer a self-contained solution without additional server components. ### [Manifold](https://mcpcatalogs.com/en/server/garume-manifold) by Garume · score 45 · ★45 Manifold is a .NET framework for defining operations once and exposing them via both CLI and MCP surfaces with high performance. **When to choose:** Choose Manifold when building .NET applications that need both CLI interfaces and MCP tool support, especially when performance is critical and you want to maintain a single operation definition. **When NOT to choose:** Avoid if you're not using .NET/C# or if you need transport-level features out-of-the-box since Man intentionally doesn't include transport implementations. ### [mcp-discovery](https://mcpcatalogs.com/en/server/rust-mcp-stack-mcp-discovery) by rust-mcp-stack · score 45 · ★83 A Rust CLI tool for discovering and documenting MCP Server capabilities with multiple output formats. **When to choose:** Choose this when you need to document MCP servers and maintain up-to-date documentation in multiple formats. **When NOT to choose:** Not ideal if you need to directly interact with MCP servers rather than just document them, or if you need runtime MCP server functionality. ### [agent-lsp](https://mcpcatalogs.com/en/server/blackwell-systems-agent-lsp) by blackwell-systems · score 45 · ★26 Stateful MCP server bridging language servers with AI agents, offering type-aware navigation, blast-radius analysis, and pre-verified edits across 30+ languages. **When to choose:** Choose agent-lsp when working with large codebases where type-aware navigation, impact analysis, and safe refactoring are critical. It's ideal for teams that need to maintain code quality across multiple programming languages. **When NOT to choose:** Avoid agent-lsp if you're working on small projects where the overhead of a persistent daemon isn't justified, or if you need support for languages not in its 30-language verified list. ### [godot-devtool](https://mcpcatalogs.com/en/server/wangdiandao-godot-devtool) by wangdiandao · score 45 · ★42 Godot 4 MCP server for AI-assisted project inspection, editing, validation, and runtime automation. **When to choose:** Choose godot-devtool if you're using Godot 4 and want AI assistants to help with game development tasks like debugging, scene editing, and runtime automation. **When NOT to choose:** Avoid if you're not using Godot 4, as this server is specifically designed for Godot game development and won't work with other engines. ### [N8N2MCP](https://mcpcatalogs.com/en/server/super-chain-n8n2mcp) by Super-Chain · score 45 · ★129 Convert N8N workflows into MCP servers for use in AI assistants like Claude and Cursor. **When to choose:** Choose N8N2MCP if you already use N8N for workflow automation and want to integrate those workflows directly into AI assistants without complex API integrations. **When NOT to choose:** Don't choose N8N2MCP if you need a completely self-contained solution without external dependencies or if you're not already invested in the N8N ecosystem. ### [metashape-mcp](https://mcpcatalogs.com/en/server/jenkinsm13-metashape-mcp) by jenkinsm13 · score 45 · ★21 MCP server providing 106 tools for Agisoft Metashape photogrammetry automation through natural language. **When to choose:** Choose this MCP server if you're already using Agisoft Metashape for photogrammetry and want to automate your processing workflows through natural language commands. **When NOT to choose:** Don't choose this if you don't have a Metashape Professional license, as the server is tightly coupled to this specific software. ### [mcp-graph-workflow](https://mcpcatalogs.com/en/server/diegonogueiradev-mcp-graph-workflow) by DiegoNogueiraDev · score 45 · ★41 MCP server that transforms PRD documents into persistent task graphs for agentic workflows. **When to choose:** Choose this when developing AI agents that need persistent task management and validation based on PRDs, especially when working offline or with strict privacy requirements. **When NOT to choose:** Avoid this if you need cloud-based collaboration features, as it's designed for local-first workflows with zero cloud dependencies. ### [maven-decoder-mcp](https://mcpcatalogs.com/en/server/salitaba-maven-decoder-mcp) by salitaba · score 45 · ★18 An MCP server for analyzing Maven jar files, extracting source code, and resolving dependencies in Java projects. **When to choose:** Choose this MCP server when working with Java projects and need AI agents to understand, search, and analyze Maven dependencies and their source code. **When NOT to choose:** Not suitable for non-Java projects or when you need to analyze dependencies in remote repositories (only works with local ~/.m2 repository). ### [unreal-mcp](https://mcpcatalogs.com/en/server/chongdashu-unreal-mcp) by chongdashu · score 45 · ★1,879 MCP server enabling AI assistants to control Unreal Engine through natural language commands. **When to choose:** Choose this if you're working with Unreal Engine and want to leverage AI assistants for automating scene setup, blueprint development, and editor control through natural language commands. **When NOT to choose:** Don't choose this if you need stable APIs for production or if you're not using Unreal Engine, as it's currently experimental and requires Unreal Engine integration. ### [kirby-mcp](https://mcpcatalogs.com/en/server/bnomei-kirby-mcp) by bnomei · score 45 · ★50 A CLI-first MCP server for Kirby CMS projects that allows inspection of blueprints/templates/plugins and interaction with a real Kirby runtime. **When to choose:** Choose Kirby MCP if you're working with Kirby CMS projects and want AI assistants to deeply understand and interact with your project structure, content, and runtime. **When NOT to choose:** Don't choose this if you're not using Kirby CMS or prefer solutions with broader CMS support, as it's specialized specifically for Kirby projects. ### [lc2mcp](https://mcpcatalogs.com/en/server/xiaotonng-lc2mcp) by xiaotonng · score 45 · ★69 A lightweight adapter that converts existing LangChain tools into FastMCP tools in one line of code. **When to choose:** When you have existing LangChain tools and want to expose them via MCP without rewriting, or when you need access to LangChain's extensive ecosystem of community tools. **When NOT to choose:** If you're building from scratch without existing LangChain tools, or if you need the opposite direction (MCP to LangChain conversion), in which case langchain-mcp-adapters would be more appropriate. ### [anythingmcp](https://mcpcatalogs.com/en/server/helpcode-ai-anythingmcp) by HelpCode-ai · score 45 · ★56 Self-hosted MCP gateway connecting REST, SOAP/WSDL, GraphQL & SQL APIs to Claude, ChatGPT, Gemini, Copilot, Cursor. **When to choose:** Choose this when you need to connect legacy systems (SOAP/WSDL, on-prem databases) to AI agents or require enterprise features like audit logging and RBAC in your MCP infrastructure. **When NOT to choose:** Don't choose this if you need write access to databases (currently read-only) or prefer a fully open-source license (BSL-1.1 with conversion to Apache 2.0 in 2030). ### [agentic-engineering-for-apache-kafka](https://mcpcatalogs.com/en/server/lensesio-agentic-engineering-for-apache-kafka) by lensesio · score 45 · ★24 Agent skills for building, operating and troubleshooting Apache Kafka applications through MCP integration. **When to choose:** Choose this if you're working with Apache Kafka and want AI agents to understand Kafka-specific best practices and perform targeted audits of your cluster configuration, consumer lag, schemas, security, and connectors. **When NOT to choose:** Don't choose this if you're not using Apache Kafka, as these skills are specifically tailored for Kafka environments and won't provide value for other messaging systems or databases. ### [MCP-Airflow-API](https://mcpcatalogs.com/en/server/call518-mcp-airflow-api) by call518 · score 45 · ★46 MCP server enabling natural language control of Apache Airflow workflows through Model Context Protocol. **When to choose:** Choose this when you want natural language control over Airflow workflows without learning complex API syntax, especially if you're managing both Airflow 2.x and 3.x environments. **When NOT to choose:** Don't choose this if you need fine-grained control over Airflow configuration beyond what's exposed through its API, or if you're using a different workflow orchestration system. ### [mcp-api-gateway](https://mcpcatalogs.com/en/server/rflpazini-mcp-api-gateway) by rflpazini · score 45 · ★40 Universal MCP server that converts any API with Swagger/OpenAPI specs into Claude Desktop tools via Docker. **When to choose:** Choose this when you need to quickly integrate multiple APIs with Claude without writing custom MCP servers, especially for REST APIs with OpenAPI specifications. **When NOT to choose:** Don't choose this if you need WebSocket support, OAuth authentication, or complex API transformations beyond what OpenAPI specs provide. ### [mcpforge](https://mcpcatalogs.com/en/server/lorenzosaraiva-mcpforge) by lorenzosaraiva · score 45 · ★17 MCPForge generates optimized MCP servers from OpenAPI specs or documentation with AI curation and workflow planning. **When to choose:** Choose MCPForge when you need to convert complex API specifications into production-ready MCP servers with curated toolsets and task-oriented workflows. **When NOT to choose:** Avoid MCPForge if you need simple endpoint wrappers without optimization or workflow planning, or if you're working with APIs that don't have clear documentation. ### [fossil-mcp](https://mcpcatalogs.com/en/server/yfedoseev-fossil-mcp) by yfedoseev · score 45 · ★56 Fossil MCP is a code quality toolkit that detects dead code, clones, and scaffolding in vibe-coded projects across 16 languages. **When to choose:** Choose this if your development workflow relies heavily on AI coding assistants and you need to detect code quality issues that traditional linters miss in AI-generated code. **When NOT to choose:** Don't choose this if you need a general-purpose code quality tool or if your development workflow doesn't involve AI coding assistants. ### [touchdesigner-mcp-server](https://mcpcatalogs.com/en/server/bottobot-touchdesigner-mcp-server) by bottobot · score 45 · ★55 Comprehensive MCP server for TouchDesigner documentation, tutorials, and Python API with 629 operators and 69 Python classes. **When to choose:** Choose this if you're working with TouchDesigner and need AI assistance with documentation lookup, Python API references, or tutorial guidance during your creative coding workflow. **When NOT to choose:** Avoid if you need write capabilities to modify TouchDesigner (this is documentation-only), or if you're working with other creative coding platforms besides TouchDesigner. ### [Ctxo](https://mcpcatalogs.com/en/server/alperhankendi-ctxo) by alperhankendi · score 45 · ★52 Ctxo is an MCP server that provides code intelligence for AI agents, delivering dependency graphs, git intent, and blast radius analysis in a single call. **When to choose:** Choose Ctxo when you need comprehensive code intelligence for AI agents working in complex codebases, particularly when analyzing dependencies and code impact is critical. **When NOT to choose:** Don't choose Ctxo if you need support for languages other than TypeScript, Go, and C#, or if you prefer minimal dependencies in your workflow. ### [orionbelt-semantic-layer](https://mcpcatalogs.com/en/server/ralfbecher-orionbelt-semantic-layer) by ralfbecher · score 45 · ★46 OrionBelt Semantic Layer converts YAML semantic models to optimized SQL for multiple databases via MCP API. **When to choose:** Teams working with multiple database dialects who want to define analytics semantics once and compile to optimized SQL across platforms through a programmatic interface. **When NOT to choose:** You need write access to databases (this is read-only only), require complex joins beyond star schema patterns, or need extensive custom SQL logic beyond the semantic model capabilities. ### [kimi-code-mcp](https://mcpcatalogs.com/en/server/howardpen9-kimi-code-mcp) by howardpen9 · score 45 · ★49 MCP server connecting Claude Code with Kimi K2.5 (256K context) to reduce token costs for codebase analysis. **When to choose:** Choose this when you're already using Kimi Code and need to analyze large codebases while minimizing expensive Claude token usage. **When NOT to choose:** Don't choose this if you don't have or want a Kimi Code subscription, or if you need read/write access to repositories (this is read-only via Kimi's CLI). ### [nvim-mcp](https://mcpcatalogs.com/en/server/paulburgess1357-nvim-mcp) by paulburgess1357 · score 45 · ★54 MCP server connecting AI agents to Neovim via msgpack-RPC, enabling code editing and command execution without plugins. **When to choose:** Choose nvim-mcp when you want AI agents to interact directly with your Neovim editor for code editing, analysis, or automation without requiring any plugins. **When NOT to choose:** Don't choose nvim-mcp if you're using a different editor, need write operations to disk without confirmation, or require authentication for secure multi-user environments. ### [Unity-AI-ProBuilder](https://mcpcatalogs.com/en/server/ivanmurzak-unity-ai-probuilder) by IvanMurzak · score 45 · ★41 Unity MCP server that enables AI assistants to create and manipulate 3D meshes through natural language commands. **When to choose:** Choose this if you're working with Unity and want AI assistants to create and manipulate 3D geometry through natural language commands. **When NOT to choose:** Not recommended if you're not using Unity or if you need high-performance mesh operations beyond basic modeling tasks. ### [mcp-ssh](https://mcpcatalogs.com/en/server/shuakami-mcp-ssh) by shuakami · score 45 · ★62 An MCP server that provides AI-powered SSH management capabilities through standardized protocols. **When to choose:** Choose this server when you need AI-powered SSH management capabilities with tmux integration and intelligent command processing. **When NOT to choose:** Avoid this if you need a pure SSH client without AI capabilities or if you're working in a restricted environment without Python support. ### [conductor-tasks](https://mcpcatalogs.com/en/server/hridaya423-conductor-tasks) by hridaya423 · score 45 · ★74 AI-powered task management system for developers that integrates via MCP or CLI. **When to choose:** Choose Conductor Tasks if you need an AI-powered task management system that supports multiple LLM providers and works both as an MCP server and standalone CLI tool. **When NOT to choose:** Don't choose this if you need a simple task manager without AI capabilities or if you prefer a single-vendor solution with extensive customization options. ### [shinzo-ts](https://mcpcatalogs.com/en/server/shinzo-labs-shinzo-ts) by shinzo-labs · score 45 · ★67 TypeScript SDK providing observability for MCP servers through OpenTelemetry integration. **When to choose:** Choose this when you need comprehensive observability for production MCP servers and are already using or planning to use OpenTelemetry for monitoring infrastructure. **When NOT to choose:** Avoid if you're looking for a lightweight solution or if you need vendor-agnostic monitoring without any vendor-specific integrations. ### [MiniMax-Coding-Plan-MCP](https://mcpcatalogs.com/en/server/minimax-ai-minimax-coding-plan-mcp) by MiniMax-AI · score 45 · ★79 Specialized MCP server providing coding-specific tools like web search and image analysis for MiniMax users. **When to choose:** Choose this MCP server if you're already using MiniMax's coding plan services and need specialized tools for web search and image analysis integrated directly into your development environment. **When NOT to choose:** Avoid this if you're not using MiniMax's API or if you need cross-platform compatibility beyond MCP clients like Claude Desktop and Cursor. ### [odoo-claude-mcp](https://mcpcatalogs.com/en/server/rosenvladimirov-odoo-claude-mcp) by rosenvladimirov · score 45 · ★18 Production-grade MCP server connecting Claude to Odoo 15-19 with 197+ tools, multi-tenant support, and Bulgaria localization. **When to choose:** Choose this if you need a production-ready, multi-tenant MCP solution for Odoo that integrates with GitHub, filesystem, and other tools, especially if you're operating in Bulgaria and need fiscal compliance features. **When NOT to choose:** Avoid if you need a simple single-user connection to Odoo, are not using Odoo 15-19, or prefer a cloud-hosted solution rather than self-hosting. ### [hypertool-mcp](https://mcpcatalogs.com/en/server/toolprint-hypertool-mcp) by toolprint · score 45 · ★153 HyperTool dynamically exposes tools from proxied MCP servers based on agent personas to help AI stay focused with relevant toolsets. **When to choose:** Choose HyperTool when working with multiple MCP servers and you need to dynamically switch between focused toolsets without restarting the AI session. **When NOT to choose:** Don't choose HyperTool if you only need a few tools from a single MCP server or if you prefer direct server connections without the added abstraction layer. ### [Alph](https://mcpcatalogs.com/en/server/aqualia-alph) by Aqualia · score 45 · ★65 Universal MCP Server Configuration Manager that automates setup across multiple AI agents. **When to choose:** When you need to configure multiple AI agents to connect to MCP servers and want a safe, automated way to manage configurations without manual file editing. **When NOT to choose:** If you only need to configure one AI agent or prefer manually editing configuration files, as Alph adds abstraction for a problem that might be simpler for single-agent setups. ### [dotbot](https://mcpcatalogs.com/en/server/andresharpe-dotbot) by andresharpe · score 45 · ★50 Zero-dependency MCP server for structured, auditable AI-assisted development in teams. **When to choose:** Choose dotbot if your team needs structured, auditable AI development workflows with multi-provider support and enterprise-grade governance. **When NOT to choose:** Avoid dotbot if you need a lightweight tool with minimal dependencies or if you prefer a language other than PowerShell for your development workflows. ### [hop](https://mcpcatalogs.com/en/server/danmartuszewski-hop) by danmartuszewski · score 45 · ★47 An elegant SSH connection manager with TUI dashboard and MCP server for AI assistants. **When to choose:** Choose Hop when you need an SSH connection manager with AI integration and a terminal dashboard for managing multiple environments across projects. **When NOT to choose:** Don't choose Hop if you need advanced SSH features not covered by the tool or if you prefer web-based UI interfaces over terminal dashboards. ### [bambu-printer-mcp](https://mcpcatalogs.com/en/server/dmontgomery40-bambu-printer-mcp) by DMontgomery40 · score 45 · ★41 A focused MCP server for Bambu Lab 3D printers offering STL manipulation, slicing, and direct printer control via MCP protocol. **When to choose:** Choose this server if you exclusively use Bambu Lab printers and want a streamlined, focused implementation without the overhead of supporting multiple printer types. **When NOT to choose:** Avoid this if you need multi-brand printer support (OctoPrint, Klipper, etc.) or if you require open-source licensing as this project is GPL-2.0 licensed. ### [cursor-notebook-mcp](https://mcpcatalogs.com/en/server/jbeno-cursor-notebook-mcp) by jbeno · score 45 · ★157 MCP server enabling AI agents to interact with Jupyter Notebook files through cell manipulation and metadata editing. **When to choose:** Choose this MCP server when working with Jupyter notebooks and you need AI agents to perform comprehensive notebook manipulation operations through MCP. **When NOT to choose:** Avoid using this if you need advanced plotting or visualization tools not provided by the underlying Jupyter ecosystem, or if you're looking for a solution with more integrated AI/notebook features beyond basic cell manipulation. ### [RevitMCPBridge2026](https://mcpcatalogs.com/en/server/weberg619-revitmcpbridge2026) by WeberG619 · score 45 · ★18 RevitMCPBridge2026 provides 705+ MCP endpoints for full AI access to Autodesk Revit via named pipes, enabling direct model manipulation and autonomous goal execution. **When to choose:** Choose RevitMCPBridge if you need comprehensive programmatic access to Revit models for AI automation, especially for large-scale modifications, documentation generation, or autonomous goal execution in AEC workflows. **When NOT to choose:** Don't choose this if you're not using Revit 2025/2026, need cross-platform compatibility (Windows-only), or require simpler read-only access to BIM data. ### [agentmako](https://mcpcatalogs.com/en/server/drhalto-agentmako) by drhalto · score 45 · ★41 Local-first MCP server providing codebase intelligence through structured context packets, diagnostics and SQLite-backed facts. **When to choose:** Choose agentmako when you need comprehensive codebase intelligence with structured context packets, especially for projects with complex routing patterns, authentication systems, or database schemas. **When NOT to choose:** Avoid agentmako if you only need basic file search without context, or if you're unwilling to commit to its local-first approach with SQLite dependency. ### [nav2_mcp_server](https://mcpcatalogs.com/en/server/ajtudela-nav2-mcp-server) by ajtudela · score 45 · ★73 An MCP server for controlling and monitoring ROS 2 Nav2 robots through navigation commands and system management. **When to choose:** Choose this server if you need AI control over Nav2-enabled robots with precise navigation commands and real-time monitoring capabilities. **When NOT to choose:** Avoid if you're not using ROS 2 or Nav2, as this server provides no functionality outside of the Nav2 ecosystem. ### [nvim-gemini-companion](https://mcpcatalogs.com/en/server/gutsavgupta-nvim-gemini-companion) by gutsavgupta · score 45 · ★82 Neovim plugin integrating Gemini and Qwen AI agents with MCP backend for seamless code assistance. **When to choose:** Choose this plugin if you're a Neovim user looking for integrated AI assistance with Gemini or Qwen, especially if you need persistent tmux sessions and diff management. **When NOT to choose:** Avoid if you're not using Neovim or if you prefer other AI tools that offer deeper IDE integration beyond what terminal-based agents can provide. ### [claude-faf-mcp](https://mcpcatalogs.com/en/server/wolfe-jam-claude-faf-mcp) by Wolfe-Jam · score 45 · ★17 FAF MCP server for Claude — persistent project context with 33 tools for better AI development. **When to choose:** Choose claude-faf-mcp when you need persistent project context for Claude across multiple sessions, especially for complex projects where maintaining context manually is impractical. **When NOT to choose:** Avoid if you're working with a different AI assistant (not Claude) or if you prefer not to use the .faf file format for project context management. ### [apra-fleet](https://mcpcatalogs.com/en/server/apra-labs-apra-fleet) by Apra-Labs · score 45 · ★39 An MCP server that orchestrates AI coding agents across machines with doer-reviewer loops and multi-machine coordination. **When to choose:** Choose Apra Fleet when you need distributed AI coding across multiple machines with built-in code review through the doer-reviewer pattern. **When NOT to choose:** Avoid if you're looking for a simple local coding tool or if you need more specialized AI features beyond code generation and review. ### [codebase-context](https://mcpcatalogs.com/en/server/patricksys-codebase-context) by PatrickSys · score 45 · ★44 An MCP server that maps codebase conventions and provides semantic search for AI agents to understand architecture and patterns. **When to choose:** Teams wanting to help AI assistants understand their specific codebase conventions and architecture before making changes, reducing the need for broad searches. **When NOT to choose:** You need write access capabilities (this tool is primarily read-only), or if you're working with very small codebases where the indexing overhead isn't justified. ### [template-mcp-server](https://mcpcatalogs.com/en/server/mcpdotdirect-template-mcp-server) by mcpdotdirect · score 45 · ★74 A starter template for building MCP servers with FastMCP supporting both stdio and HTTP transports. **When to choose:** Choose this template if you're a TypeScript developer looking to quickly scaffold a new MCP server with both stdio and HTTP transport options. **When NOT to choose:** Don't choose this if you need server-side authentication, as the template doesn't include an authentication model. ### [laravel-mcp-companion](https://mcpcatalogs.com/en/server/brianirish-laravel-mcp-companion) by brianirish · score 45 · ★34 An MCP server providing comprehensive Laravel documentation aggregation and navigation with multi-version support, learning paths, and package discovery. **When to choose:** Choose this server when you need comprehensive Laravel ecosystem documentation with enhanced search, learning paths, and cross-package compatibility for development or reference purposes. **When NOT to choose:** Don't choose this if you need project-aware context code generation (use Laravel Boost instead) or documentation for non-Laravel libraries (use Context7). ### [mcp-mermaid-validator](https://mcpcatalogs.com/en/server/rtuin-mcp-mermaid-validator) by rtuin · score 45 · ★54 MCP server for validating and rendering Mermaid diagrams into PNG format. **When to choose:** Choose this MCP server when your application needs to validate and render Mermaid diagrams as PNG images in an AI workflow. **When NOT to choose:** Don't choose this if you need SVG output format or require more advanced diagram manipulation features beyond validation and basic rendering. ### [mcp-link](https://mcpcatalogs.com/en/server/automation-ai-labs-mcp-link) by automation-ai-labs · score 45 · ★605 MCP Link automatically converts any OpenAPI V3 API into a fully-functional MCP server. **When to choose:** Choose MCP Link when you need to quickly convert existing REST APIs with OpenAPI specs into MCP servers without manual implementation. **When NOT to choose:** Don't choose this if you need advanced OAuth flows, resource-based API interactions, or MIME type handling beyond what's currently supported. ### [mcpx](https://mcpcatalogs.com/en/server/lydakis-mcpx) by lydakis · score 45 · ★52 mcpx is a CLI tool that composes MCP servers into command-line interfaces. **When to choose:** Choose mcpx when you're a CLI power user who wants to integrate MCP tools into existing shell workflows and pipelines, or when you need to quickly test MCP servers before implementing them in larger systems. **When NOT to choose:** Don't choose mcpx if you primarily work through GUI interfaces, or if you need deep integration with specific AI agents beyond basic skill injection. ### [mTarsier](https://mcpcatalogs.com/en/server/mcp360-mtarsier) by mcp360 · score 45 · ★38 mTarsier is an MCP server management platform that centralizes and simplifies MCP server configuration across multiple AI clients. **When to choose:** Choose mTarsier when you need a unified solution to manage MCP servers and skills across multiple AI clients like Claude, Cursor, and VS Code without switching between different configuration files. **When NOT to choose:** Avoid mTarsier if you're looking for a lightweight tool with minimal dependencies or if you only need to manage MCP servers for a single client like only Claude Desktop. ### [servicenow-mcp](https://mcpcatalogs.com/en/server/aartiq-servicenow-mcp) by aartiq · score 45 · ★23 Comprehensive MCP server for ServiceNow with 400+ tools across all modules, supporting OAuth authentication and role-based tool packages. **When to choose:** Choose this if you need comprehensive ServiceNow integration with AI assistants and require production-ready security with role-based access controls. **When NOT to choose:** Not suitable if you're not using ServiceNow, as this is specifically designed for ServiceNow integration only. ### [time-mcp](https://mcpcatalogs.com/en/server/yokingma-time-mcp) by yokingma · score 45 · ★66 Time MCP server providing time awareness tools to LLMs with timezone conversion, current time, and date calculations. **When to choose:** Choose this server when your application needs precise time calculations and timezone conversions, especially when working with scheduling applications or systems that operate across multiple time zones. **When NOT to choose:** Don't choose this if you need historical time data, calendar management features, or complex date arithmetic beyond what's listed in the tools. ### [mcp-server-semgrep](https://mcpcatalogs.com/en/server/vetcoders-mcp-server-semgrep) by vetcoders · score 44 · ★28 MCP Server Semgrep integrates Semgrep static analysis with AI assistants for code scanning, security detection, and quality improvements. **When to choose:** Choose this server for teams already using Semgrep who want to integrate static analysis directly into their AI workflow for code quality and security auditing. **When NOT to choose:** Not suitable for environments where you can't install Semgrep CLI or need scanning capabilities without external dependencies. ### [ILSpy-Mcp](https://mcpcatalogs.com/en/server/bivex-ilspy-mcp) by bivex · score 44 · ★33 A .NET assembly decompiler and analysis tool that exposes ILSpy's capabilities through the MCP protocol for AI assistants. **When to choose:** Choose ILSpy-MCP when you need to reverse engineer .NET assemblies through AI assistants without leaving your development environment. **When NOT to choose:** Don't choose this if you need to analyze non-.NET code or require write access to assemblies, as this server is read-only and limited to the .NET ecosystem. ### [minecraft-dev-mcp](https://mcpcatalogs.com/en/server/mcdxai-minecraft-dev-mcp) by MCDxAI · score 44 · ★17 MCP server providing 20 tools for Minecraft mod development including decompilation, mapping, and source code analysis. **When to choose:** Choose this MCP server when working with Minecraft mod development and you need AI to have direct access to decompiled source code, mapping systems, and validation tools without manual setup. **When NOT to choose:** Don't choose this if you're not working with Minecraft mod development, as the tools are specialized for this ecosystem and won't be useful for other projects. ### [gospy](https://mcpcatalogs.com/en/server/monsterxx03-gospy) by monsterxx03 · score 44 · ★96 Go process inspector MCP server with goroutine analysis, memory statistics, and binary information via terminal UI or API. **When to choose:** Choose gospy when you need to analyze running Go applications without restarting them or modifying source code, particularly for debugging goroutine issues or monitoring memory usage in production environments. **When NOT to choose:** Don't choose gospy if you need to inspect non-Go processes, don't have root privileges, or need cross-platform support beyond Linux and macOS. ### [agentify](https://mcpcatalogs.com/en/server/monadworks-agentify) by MonadWorks · score 44 · ★25 Converts OpenAPI specs to MCP servers and other AI agent interfaces with a single command. **When to choose:** Choose Agentify when you need to quickly convert existing OpenAPI APIs into MCP servers and multiple AI agent formats without manual implementation. **When NOT to choose:** Don't choose Agentify if you need highly customized MCP server behavior or if your API specifications contain complex security patterns that can't be auto-detected. ### [ontomics](https://mcpcatalogs.com/en/server/etiennechollet-ontomics) by EtienneChollet · score 44 · ★31 ontomics extracts domain knowledge from codebases via MCP, reducing token consumption 20x and search time 10x by making concepts and naming conventions queryable. **When to choose:** Choose ontomics when working with large codebases where you need to quickly understand domain concepts, naming conventions, and behavioral relationships without excessive token consumption. **When NOT to choose:** Avoid ontomics if you need real-time collaboration features or if your project uses languages not supported by the tool (currently only Python, TypeScript, JavaScript, and Rust). ### [zed-mcp-server-context7](https://mcpcatalogs.com/en/server/akbxr-zed-mcp-server-context7) by akbxr · score 44 · ★106 MCP server that integrates Context7 to fetch up-to-date documentation for Zed Assistant. **When to choose:** Choose this MCP server if you're using Zed and need up-to-date documentation for modern frameworks, especially when working with Next.js, React, or other rapidly evolving technologies. **When NOT to choose:** Don't choose this if you're not using Zed or if you need documentation for niche frameworks not supported by Context7 service. ### [holoviz-mcp](https://mcpcatalogs.com/en/server/marcskovmadsen-holoviz-mcp) by MarcSkovMadsen · score 44 · ★33 An MCP server providing intelligent access to HoloViz ecosystem for building data visualizations and dashboards. **When to choose:** Choose this when working with HoloViz ecosystem libraries and need AI assistance for data visualization and dashboard development. **When NOT to choose:** Avoid if security is a concern due to Python code execution capabilities, or if you need visualization support for libraries outside the HoloViz ecosystem. ### [1xn-vmcp](https://mcpcatalogs.com/en/server/1xn-labs-1xn-vmcp) by 1xn-labs · score 44 · ★49 vMCP is a virtual MCP server platform that combines multiple MCP servers into a unified endpoint with customization capabilities. **When to choose:** Choose vMCP when you need to manage multiple MCP servers across different AI clients (Claude, ChatGPT, Cursor) and want unified authentication, tool composition, and customization without code changes. **When NOT to choose:** Avoid vMCP if you only need a single MCP server, as it adds complexity that's unnecessary for simple use cases. ### [lyra-tool-discovery](https://mcpcatalogs.com/en/server/nirholas-lyra-tool-discovery) by nirholas · score 44 · ★25 AI-powered discovery tool that finds and analyzes crypto-focused MCP servers across GitHub and npm, generating ready-to-use plugin configurations. **When to choose:** Choose Lyra when you need to discover and evaluate blockchain-related MCP servers and APIs for crypto applications, especially when you want to automate the evaluation and configuration generation process. **When NOT to choose:** Don't choose Lyra if you're working outside the blockchain/crypto ecosystem, as its specialized filtering and keyword analysis are optimized for crypto-related tools and protocols. ### [mcpgen](https://mcpcatalogs.com/en/server/lyeslabs-mcpgen) by lyeslabs · score 44 · ★90 mcpgen is a Go tool that generates MCP server boilerplate from OpenAPI 3 specifications. **When to choose:** Choose mcpgen when you have existing OpenAPI documentation and want to rapidly expose those APIs as MCP tools without manually implementing server scaffolding. **When NOT to choose:** Don't choose mcpgen if you need to support protocols other than OpenAPI or if you're working with APIs that can't be described with OpenAPI specifications. ### [symfony-mcp-server](https://mcpcatalogs.com/en/server/klapaudius-symfony-mcp-server) by klapaudius · score 44 · ★30 A Symfony package for building enterprise-grade MCP servers with real-time communication and AI agent capabilities. **When to choose:** Choose this when building enterprise-grade AI agents within Symfony applications that require secure real-time communication and intelligent tool behavior with LLM integration. **When NOT to choose:** Avoid if you're not using Symfony or PHP in your tech stack, as this is tightly coupled to the Symfony framework and won't integrate with other backend technologies. ### [booklib](https://mcpcatalogs.com/en/server/booklib-ai-booklib) by booklib-ai · score 44 · ★28 BookLib is an MCP server that detects AI knowledge gaps in projects and injects context at runtime to fix outdated code patterns. **When to choose:** Choose BookLib when your team uses multiple AI coding assistants and needs to ensure consistent code patterns across libraries released after AI training cutoffs. **When NOT to choose:** Don't choose BookLib if your project doesn't use post-training dependencies or if you prefer a simple, single-tool solution without MCP integration. ### [asc-mcp](https://mcpcatalogs.com/en/server/zelentsov-dev-asc-mcp) by zelentsov-dev · score 44 · ★27 MCP server providing 208 tools for managing App Store Connect resources directly from Claude or other MCP clients. **When to choose:** Choose this MCP server if you need comprehensive automation of iOS/macOS app releases through natural language, especially when managing multiple apps or TestFlight beta programs. **When NOT to choose:** Avoid if you're not on macOS (requires 14.0+) or if you primarily manage Android apps; also consider alternatives if you only need basic App Store Connect functionality without the extensive toolset. ### [Unreal-MCP](https://mcpcatalogs.com/en/server/aadeshrao123-unreal-mcp) by aadeshrao123 · score 44 · ★28 MCP server providing 280+ commands to control Unreal Engine 5 from AI coding assistants. **When to choose:** Choose this if you work with Unreal Engine 5 and want to integrate AI assistants directly into your development workflow for asset creation and management. **When NOT to choose:** Don't choose this if you don't use Unreal Engine 5, or if you prefer a more general-purpose game engine interface. ### [mcp-jfrog](https://mcpcatalogs.com/en/server/jfrog-mcp-jfrog) by jfrog · score 44 · ★118 Experimental MCP server for JFrog platform API providing repository management, build tracking, and security scanning capabilities. **When to choose:** Choose this server if you're already using JFrog for artifact management and want AI assistants to help with repository operations, build tracking, and security monitoring. **When NOT to choose:** Avoid this if you need production stability as it's marked as experimental, or if you're not already invested in the JFrog ecosystem. ### [reporecall](https://mcpcatalogs.com/en/server/proofofwork-agency-reporecall) by proofofwork-agency · score 44 · ★30 Local codebase memory system with AST indexing, call graphs, and business context generation for Claude Code and MCP agents. **When to choose:** Choose Reporecall when working with large codebases that need intelligent context retrieval, architecture visualization, and want to reduce token usage in AI-assisted coding without cloud dependencies. **When NOT to choose:** Don't choose Reporecall if you need frequent write access to the codebase through MCP tools, as this appears to be primarily a read-only system focused on context retrieval. ### [testrail-mcp-server](https://mcpcatalogs.com/en/server/uarlouski-testrail-mcp-server) by uarlouski · score 44 · ★25 MCP server connecting AI assistants to TestRail for test case management and execution through natural language. **When to choose:** Choose this if you're already using TestRail and want to integrate AI assistants directly into your QA workflow without context switching. **When NOT to choose:** Don't choose this if you're not already using TestRail, as it only integrates with this specific test management platform. ### [bzm-mcp](https://mcpcatalogs.com/en/server/blazemeter-bzm-mcp) by Blazemeter · score 44 · ★25 BlazeMeter MCP server enables AI-driven performance testing via natural language commands. **When to choose:** Choose this if you're already using BlazeMeter for performance testing and want to automate test workflows through AI assistants. **When NOT to choose:** Avoid if you need a self-hosted solution or are concerned about vendor lock-in to BlazeMeter's platform. ### [rtfmbro-mcp](https://mcpcatalogs.com/en/server/marckrenn-rtfmbro-mcp) by marckrenn · score 44 · ★87 rtfmbro provides always-up-to-date, version-specific package documentation as context for coding agents via MCP. **When to choose:** Choose rtfmbro when you need version-specific documentation for packages with frequent updates or when working with legacy projects that use older package versions. **When NOT to choose:** Avoid rtfmbro if you primarily work with languages not yet supported (Rust, Go, Java, etc.) or if you need access to private repositories (not yet supported). ### [cli](https://mcpcatalogs.com/en/server/mcpgod-cli) by mcpgod · score 44 · ★115 MCPGod is a CLI tool for managing MCP servers with fine-grained control over clients, tools, and permissions. **When to choose:** Choose MCPGod when you need granular control over MCP server management across multiple clients with tool-level permissions and detailed logging. **When NOT to choose:** Avoid if you need direct server implementation or a GUI interface, as MCPGod is purely a command-line management tool. ### [unreal-mcp](https://mcpcatalogs.com/en/server/runreal-unreal-mcp) by runreal · score 44 · ★101 An MCP server for Unreal Engine that uses Python remote execution without requiring a new plugin. **When to choose:** Choose this server if you need AI assistants to interact with Unreal Engine projects without installing additional plugins, especially for asset management, scene manipulation, and editor automation. **When NOT to choose:** Avoid this if you need deep integration with C++ systems or require robust security measures, as it provides full editor access with limited safeguards against unintended changes. ### [mcp-annotated-java-sdk](https://mcpcatalogs.com/en/server/thought2code-mcp-annotated-java-sdk) by thought2code · score 44 · ★29 Annotation-driven MCP Java SDK for simplified MCP server development with zero boilerplate. **When to choose:** Java developers who want to build MCP servers with minimal boilerplate code without using Spring Framework. **When NOT to choose:** Projects requiring complex authentication mechanisms or those needing to integrate with existing Spring-based architectures. ### [fastify-mcp-server](https://mcpcatalogs.com/en/server/flaviodelgrosso-fastify-mcp-server) by flaviodelgrosso · score 44 · ★27 Fastify plugin for creating MCP HTTP servers with session management, authentication, and transport options. **When to choose:** Choose this if you're building MCP servers within a Fastify-based microservices architecture and need HTTP transport with session management capabilities. **When NOT to choose:** Avoid if you need non-HTTP transport protocols or require extensive features beyond MCP's core protocol capabilities. ### [pfc-mcp](https://mcpcatalogs.com/en/server/yusong652-pfc-mcp) by yusong652 · score 44 · ★37 MCP server connecting AI agents to ITASCA PFC for running DEM simulations through natural conversation. **When to choose:** Choose pfc-mcp when working with ITASCA PFC simulations and want to leverage AI agents for documentation browsing, code execution, and simulation management through natural conversation. **When NOT to choose:** Not suitable if you don't have access to ITASCA PFC software, or need a solution for non-PFC DEM simulation tools. ### [Unity-AI-ParticleSystem](https://mcpcatalogs.com/en/server/ivanmurzak-unity-ai-particlesystem) by IvanMurzak · score 44 · ★36 AI-powered Unity Particle System manipulation through natural language commands via MCP protocol. **When to choose:** Unity developers looking to streamline particle system workflows through natural language commands without manual inspector navigation. **When NOT to choose:** Not suitable for developers working with other game engines, or those who prefer traditional manual particle system configuration methods. ### [leanmcp-sdk](https://mcpcatalogs.com/en/server/leanmcp-leanmcp-sdk) by LeanMCP · score 44 · ★28 Production-ready TypeScript SDK for building enterprise-grade MCP servers with authentication, multi-tenant support, and observability. **When to choose:** Choose LeanMCP when building production MCP servers that require enterprise features like authentication, multi-tenancy, and observability. **When NOT to choose:** Avoid for simple MCP servers without enterprise requirements, as it adds significant complexity compared to basic MCP implementations. ### [cheat-engine-server-python](https://mcpcatalogs.com/en/server/bethington-cheat-engine-server-python) by bethington · score 44 · ★49 MCP Cheat Engine Server provides safe read-only memory access for debugging and analysis. **When to choose:** Choose this MCP server when you need safe, read-only memory analysis for debugging applications, security research, or game modding without risking system modifications. **When NOT to choose:** Don't choose this if you need to modify memory or if you're working on platforms other than Windows (support is limited on Linux/macOS). ### [css-noop-checker](https://mcpcatalogs.com/en/server/purupurupu-css-noop-checker) by purupurupu · score 44 · ★25 Chrome DevTools extension with MCP server for detecting CSS properties with no effect on elements. **When to choose:** Choose this server when you need AI assistance to detect CSS properties that have no effect on elements, particularly when analyzing web pages or CSS code for optimization opportunities. **When NOT to choose:** Don't choose this if you need CSS validation against the official specifications rather than browser behavior, or if you're working with non-Chromium browsers. ### [claude-code-mcp](https://mcpcatalogs.com/en/server/kunihiros-claude-code-mcp) by KunihiroS · score 44 · ★34 An MCP server that bridges less capable LLMs with Claude Code by exposing code analysis tools via stdio. **When to choose:** Choose this MCP server when you need to integrate Claude Code's analysis capabilities into MCP-compatible hosts without directly using the CLI. **When NOT to choose:** Don't choose this if you need direct CLI access to Claude Code, prefer solutions without dependency on Claude Code installation, or require more comprehensive code analysis tools. ### [sourcerer-mcp](https://mcpcatalogs.com/en/server/st3v3nmw-sourcerer-mcp) by st3v3nmw · score 44 · ★113 MCP server for semantic code search that helps AI agents navigate codebases efficiently without token waste. **When to choose:** Choose Sourcerer when working with large codebases and you need AI agents to efficiently find relevant code without reading entire files. **When NOT to choose:** Avoid if you need offline operation or have concerns about vendor lock-in to OpenAI for embeddings. ### [jenkins-mcp-enterprise](https://mcpcatalogs.com/en/server/jordan-jarvis-jenkins-mcp-enterprise) by Jordan-Jarvis · score 44 · ★28 Advanced Jenkins MCP server with enterprise features for multi-instance management and AI-powered build diagnostics. **When to choose:** Choose this server if you need advanced Jenkins integration with AI-powered failure analysis and multi-instance management capabilities. **When NOT to choose:** Avoid this if you need a simple Jenkins integration, require job editing capabilities for SCM-backed pipelines, or prefer a permissive license (this uses GPL v3). ### [mcp-expose-abilities](https://mcpcatalogs.com/en/server/bjornfix-mcp-expose-abilities) by bjornfix · score 44 · ★25 WordPress MCP server with 61+ core abilities and 12+ add-ons for content, SEO, and site management via AI agents. **When to choose:** Choose this if you're already using WordPress and want AI agents to directly manage content, plugins, and site operations without copy-paste workflows between chat and wp-admin. **When NOT to choose:** Don't choose if you need non-WordPress site management, or if you prefer not to install multiple dependent plugins (Abilities API, MCP Adapter) to get a working system. ### [mcp-python-interpreter](https://mcpcatalogs.com/en/server/yzfly-mcp-python-interpreter) by yzfly · score 44 · ★99 MCP Python interpreter server enabling LLMs to run Python code, manage environments, and handle files with proper security isolation. **When to choose:** Choose this MCP server when you need to enable AI agents to execute Python code in a controlled environment, manage Python environments and packages, or work with files within a specific directory. **When NOT to choose:** Avoid using this server if you need broader system access outside the isolated working directory or require access to environments beyond Python. ### [swiftlens](https://mcpcatalogs.com/en/server/swiftlens-swiftlens) by swiftlens · score 44 · ★124 SwiftLens MCP server enables AI models to understand Swift codebases with compiler-grade accuracy via SourceKit-LSP integration. **When to choose:** Choose SwiftLens when working with Swift codebases and need AI assistance with code analysis, refactoring, or documentation generation with compiler-level accuracy. **When NOT to choose:** Avoid if not using Swift or macOS, as SourceKit-LSP integration requires Apple's development tools exclusively. ### [kicad-studio](https://mcpcatalogs.com/en/server/oaslananka-kicad-studio) by oaslananka · score 44 · ★27 VS Code extension that integrates KiCad projects with MCP tools and AI capabilities for PCB design workflows. **When to choose:** Choose KiCad Studio if you're using KiCad for PCB design and want AI-assisted workflows integrated directly into VS Code with MCP server capabilities. **When NOT to choose:** Don't choose this if you're not using KiCad, as it's specifically designed for KiCad workflows and may not be useful for other EDA tools. ### [gh-mcp](https://mcpcatalogs.com/en/server/shuymn-gh-mcp) by shuymn · score 44 · ★29 GitHub CLI extension that runs github-mcp-server using existing gh authentication without manual PAT setup. **When to choose:** Choose this if you already use GitHub CLI and want seamless MCP integration without managing separate Personal Access Tokens. **When NOT to choose:** Not suitable if you need fine-grained control over GitHub permissions or are not already using the GitHub CLI. ### [mcp-klever-vm](https://mcpcatalogs.com/en/server/klever-io-mcp-klever-vm) by klever-io · score 44 · ★31 MCP server for Klever blockchain smart contract development with pattern extraction and validation tools. **When to choose:** Choose this MCP server when developing on the Klever blockchain and need access to a comprehensive knowledge base of contract patterns, best practices, and validation tools. **When NOT to choose:** Avoid if you're working on other blockchain platforms as this server is specifically tailored for Klever's ecosystem with limited applicability elsewhere. ### [markview](https://mcpcatalogs.com/en/server/paulhkang94-markview) by paulhkang94 · score 44 · ★29 Native macOS markdown preview with MCP server for Claude Code, featuring GFM, Mermaid, and syntax highlighting. **When to choose:** Choose MarkView when working on macOS and need a native, lightweight markdown preview integrated with Claude Code that supports GFM, Mermaid diagrams, and real-time rendering. **When NOT to choose:** Don't choose MarkView if you need cross-platform support or require advanced markdown features beyond what GFM provides, as it's macOS-only and focused on core rendering capabilities. ### [archmcp](https://mcpcatalogs.com/en/server/dejo1307-archmcp) by dejo1307 · score 44 · ★30 archmcp is an MCP server that generates compact architectural snapshots of repositories to provide AI agents with structured codebase overviews. **When to choose:** Choose archmcp when you work with multiple codebases across different languages and want your AI coding assistant to understand architectural relationships before exploring code. **When NOT to choose:** Don't choose archmcp if you only work with a single small project or need real-time code analysis as you type, as it's designed for upfront architectural analysis rather than continuous monitoring. ### [furi](https://mcpcatalogs.com/en/server/ashwwwin-furi) by ashwwwin · score 44 · ★88 A comprehensive CLI & API tool for managing MCP servers with process management, HTTP routing, and SSE aggregation capabilities. **When to choose:** Choose Furi when you need a comprehensive solution for managing multiple MCP servers with process monitoring, HTTP API exposure, and tool aggregation capabilities. **When NOT to choose:** Avoid Furi if you only need basic MCP management without process monitoring or if you're working primarily with Python-based MCP servers which are not yet fully supported. ### [esp-mcp](https://mcpcatalogs.com/en/server/horw-esp-mcp) by horw · score 44 · ★148 MCP server for ESP32 development that centralizes ESP-IDF commands and simplifies LLM-driven interaction. **When to choose:** Choose ESP-MCP if you're developing for ESP32/ESP-IDF and want LLM-driven automation of common development tasks through a centralized interface. **When NOT to choose:** Avoid if you need non-ESP32 development tools, require advanced features not yet implemented in this PoC, or prefer traditional CLI workflows over LLM integration. ### [fastify-mcp](https://mcpcatalogs.com/en/server/haroldadmin-fastify-mcp) by haroldadmin · score 44 · ★27 A Fastify plugin that enables running MCP servers with Streamable HTTP and HTTP+SSE transports. **When to choose:** Choose fastify-mcp when building Fastify-based applications that need to expose MCP functionality, especially if you need both stateful and stateless server modes with session management. **When NOT to choose:** Not recommended if you're not using Fastify as your web framework, or if you need MCP server features beyond tools and resources like advanced prompt capabilities. ### [context-harness](https://mcpcatalogs.com/en/server/parallax-labs-context-harness) by parallax-labs · score 44 · ★36 Local-first context ingestion and retrieval framework with SQLite, embeddings, and MCP server for AI tools like Cursor and Claude. **When to choose:** Choose Context Harness when you need a local-first context ingestion system with support for multiple data sources and hybrid search capabilities for AI tools. **When NOT to choose:** Avoid this if you need cloud-based context storage, require real-time collaboration features, or prefer a fully-managed service rather than a self-hosted solution. ### [claude-dev-suite](https://mcpcatalogs.com/en/server/claude-dev-suite-claude-dev-suite) by claude-dev-suite · score 44 · ★15 A comprehensive AI development toolkit extending Claude Code with specialized agents, MCP servers, and visual orchestration. **When to choose:** Teams already using Claude Code who need specialized agents, MCP tools, and visual orchestration for complex multi-agent development workflows across multiple technologies. **When NOT to choose:** Projects requiring simple, lightweight AI assistance without the overhead of multiple agents and orchestration, or those needing integration with AI assistants other than Claude. ### [mcp_on_ruby](https://mcpcatalogs.com/en/server/nagstler-mcp-on-ruby) by nagstler · score 44 · ★98 A production-ready Ruby implementation of MCP for Rails applications with authentication and tool/resource exposure. **When to choose:** Choose this MCP server if you're building on Rails and want to expose your application's functionality and data to AI agents with proper security and authentication. **When NOT to choose:** Don't choose this if you're not using Ruby/Rails or need more advanced real-time features like Server-Sent Events (SSE) which are still in development. ### [azure-diagram-mcp](https://mcpcatalogs.com/en/server/dminkovski-azure-diagram-mcp) by dminkovski · score 44 · ★25 An MCP server that converts natural language descriptions into Azure architecture diagrams using Python's Diagrams library. **When to choose:** Choose this when you need to quickly generate Azure architecture diagrams through natural language prompts without using diagramming tools. **When NOT to choose:** Don't choose this if you need detailed customization of diagrams or support for non-Azure cloud providers. ### [mcp](https://mcpcatalogs.com/en/server/posthog-mcp) by PostHog · score 44 · ★142 Official PostHog MCP server for integrating analytics data with AI applications. **When to choose:** Choose this server if you already use PostHog for analytics and want to query your product data directly from AI assistants. **When NOT to choose:** Avoid this if you don't use PostHog or need write access to your analytics data (this server provides read-only access). ### [DecisionNode](https://mcpcatalogs.com/en/server/decisionnode-decisionnode) by decisionnode · score 44 · ★29 DecisionNode is a CLI and local MCP server for structured memory storage with semantic search across multiple AI coding environments. **When to choose:** Choose DecisionNode if you need to maintain consistent technical decisions across multiple AI coding environments and want semantic search capabilities to retrieve relevant context based on meaning rather than exact keywords. **When NOT to choose:** Don't choose DecisionNode if you need real-time collaboration on decisions or if your team prefers more traditional wikis or documentation systems over a structured, queryable memory store. ### [ronykit](https://mcpcatalogs.com/en/server/clubpay-ronykit) by clubpay · score 44 · ★36 Go-based API framework with built-in MCP server for AI-assisted development scaffolding. **When to choose:** Choose RonyKit's MCP server when developing Go APIs with AI assistance, especially when you want type safety, auto-generated documentation, and consistent scaffolding across your projects. **When NOT to choose:** Don't choose RonyKit if you're not working with Go APIs, or if you prefer a different language ecosystem for your AI-assisted development. ### [ContextPods](https://mcpcatalogs.com/en/server/conorluddy-contextpods) by conorluddy · score 44 · ★38 ContextPods is a comprehensive MCP development suite that can generate and manage local MCP servers in multiple languages. **When to choose:** Choose Context-Pods when you need to rapidly create and manage multiple MCP servers across different languages, or want to convert existing scripts into MCP servers with minimal effort. **When NOT to choose:** Avoid if you're building a simple, single-purpose MCP server or need authentication/authorization features, as Context-Pods focuses on development infrastructure rather than security implementations. ### [mermaid-mcp](https://mcpcatalogs.com/en/server/narasimhaponnada-mermaid-mcp) by Narasimhaponnada · score 44 · ★42 AI-powered Mermaid diagram generation MCP server for GitHub Copilot, Claude, and custom LLM applications. **When to choose:** Choose this when you need to generate diagrams through AI assistants and want professional-quality SVG output without installing diagramming tools locally. **When NOT to choose:** Don't choose this if you need real-time collaboration features or require diagram editing capabilities beyond what AI can generate. ### [lazy-tool](https://mcpcatalogs.com/en/server/mcp-shark-lazy-tool) by mcp-shark · score 44 · ★25 A local-first MCP discovery runtime that reduces prompt bloat by searching tools before invocation. **When to choose:** Choose lazy-tool if you have multiple MCP servers and want to reduce token usage without cloud dependencies or complex infrastructure. **When NOT to choose:** Don't choose lazy-tool if you only use Claude's built-in tool search or if you prefer managed cloud solutions over local-first approaches. ### [heimdall-mcp-server](https://mcpcatalogs.com/en/server/lcbcfoo-heimdall-mcp-server) by lcbcFoo · score 44 · ★104 Heimdall MCP Server provides persistent memory for AI coding assistants by indexing documentation and git history. **When to choose:** Choose Heimdall if you want persistent memory for AI coding assistants across development sessions and need context-aware retrieval from your codebase and git history. **When NOT to choose:** Avoid Heimdall if you're working with sensitive codebases that can't be indexed externally or if you prefer a minimal setup without Docker dependencies. ### [Vibe-Coder-MCP](https://mcpcatalogs.com/en/server/freshtechbro-vibe-coder-mcp) by freshtechbro · score 44 · ★99 Vibe-Coder-MCP is a TypeScript-based MCP server that extends AI assistants with specialized software development tools. **When to choose:** Choose Vibe-Coder-MCP when you need specialized software development tools integrated directly into your AI assistant for comprehensive project development from research to implementation. **When NOT to choose:** Avoid this if you're not comfortable using OpenRouter API for AI processing or if you need tools beyond software development scope. ### [GEmojiSharp](https://mcpcatalogs.com/en/server/hlaueriksson-gemojisharp) by hlaueriksson · score 44 · ★156 GEmojiSharp.McpServer provides GitHub emoji tools for C# applications via the MCP protocol. **When to choose:** Choose this MCP server if you need GitHub emoji functionality in .NET applications or want to provide emoji lookup capabilities to AI systems. **When NOT to choose:** Avoid this if you need emoji functionality for platforms other than GitHub, as it's specifically designed for GitHub's emoji set. ### [review-flow](https://mcpcatalogs.com/en/server/dgouron-review-flow) by DGouron · score 44 · ★37 Multi-agent AI code review system with MCP integration for GitLab MRs and GitHub PRs. **When to choose:** Choose review-flow for teams already using Claude Code who want automated, multi-faceted code reviews with real-time monitoring and team insights. **When NOT to choose:** Avoid if you need support for version control systems other than GitLab/GitHub, or if you prefer not to use Claude Code as the underlying AI model. ### [codingbuddy](https://mcpcatalogs.com/en/server/jeremydev87-codingbuddy) by JeremyDev87 · score 44 · ★31 Codingbuddy is an MCP server that orchestrates 37 specialized AI agents through a PLAN → ACT → EVAL workflow to improve code quality. **When to choose:** Choose Codingbuddy when you need structured AI-driven development with measurable quality improvements across multiple coding tools and want to coordinate specialized AI agents for different aspects of software development. **When NOT to choose:** Avoid Codingbuddy if you prefer a simpler coding assistant or need a solution that works with proprietary AI models not included in the supported list. ### [elysia-mcp](https://mcpcatalogs.com/en/server/kerlos-elysia-mcp) by kerlos · score 44 · ★41 A TypeScript plugin for ElysiaJS to implement MCP servers with HTTP transport and session management. **When to choose:** Choose this when you're building MCP servers with ElysiaJS and need HTTP transport with session management. **When NOT to choose:** Avoid if you need non-HTTP transports (like SSE or WebSocket) or are not using ElysiaJS framework. ### [keryx](https://mcpcatalogs.com/en/server/actionhero-keryx) by actionhero · score 44 · ★29 Fullstack TypeScript framework where actions work as HTTP, WebSocket, CLI, tasks, and MCP tools automatically. **When to choose:** Choose Keryx when you're building a new fullstack TypeScript application and want the same business logic to work for both human users and AI agents without duplicating code. **When NOT to choose:** Don't choose Keryx if you're already invested in a Node.js ecosystem (it requires Bun), need complex database relationships beyond what Drizzle supports, or require multi-language support for your MCP server. ### [mcp-zero](https://mcpcatalogs.com/en/server/zeromicro-mcp-zero) by zeromicro · score 44 · ★42 MCP server for go-zero framework that generates APIs, RPC services, and models with AI assistance. **When to choose:** Choose this when working with go-zero framework and want to accelerate development through AI-assisted code generation. **When NOT to choose:** Don't choose this if you're not using the go-zero framework or prefer manual coding without AI assistance. ### [zed-mcp-server-github](https://mcpcatalogs.com/en/server/loamstudios-zed-mcp-server-github) by LoamStudios · score 43 · ★88 A Zed extension that integrates GitHub MCP Server as a context provider for AI assistance. **When to choose:** Choose this if you're a Zed user who wants GitHub repository context integrated into your IDE's AI workflow. **When NOT to choose:** Don't choose this if you use a different IDE besides Zed, or if you need access to GitHub features beyond repository context. ### [mcp-server-node](https://mcpcatalogs.com/en/server/lucianoayres-mcp-server-node) by lucianoayres · score 43 · ★75 A Node.js MCP server implementation with simple math tools and environment variable retrieval. **When to choose:** Choose this if you're looking for a clear Node.js example of how to implement MCP tools with validation and environment variable integration. **When NOT to choose:** Don't choose this if you need production-ready features or more complex tool implementations, as it's primarily a demonstration/example. ### [codex-mcp-tool](https://mcpcatalogs.com/en/server/x51xxx-codex-mcp-tool) by x51xxx · score 43 · ★21 MCP server connecting Claude/Cursor to Codex CLI with file analysis, multi-turn conversations, and local model support. **When to choose:** Choose this server when you're already using Codex CLI and want seamless integration with Claude/Cursor for code analysis, especially for multi-turn conversations about your codebase. **When NOT to choose:** Avoid if you need write access to files (this is read-only) or if you're not already using Codex CLI, as the setup requires additional CLI installation beyond MCP configuration. ### [trueline-mcp](https://mcpcatalogs.com/en/server/rjkaes-trueline-mcp) by rjkaes · score 43 · ★17 An MCP plugin that provides hash-verified file editing and targeted reads for AI coding agents to reduce token usage and prevent silent corruption. **When to choose:** Choose trueline-mcp when working with AI coding agents on large codebases where reducing token usage and preventing silent file corruption are priorities. **When NOT to choose:** Don't choose trueline-mcp if you primarily work with small files under 3KB where the built-in tools are sufficient, or if your AI platform doesn't support MCP hooks. ### [drf-mcp-docs](https://mcpcatalogs.com/en/server/abdulkhalek-1-drf-mcp-docs) by Abdulkhalek-1 · score 43 · ★17 An MCP server that exposes Django REST Framework API documentation to AI coding agents for frontend development assistance. **When to choose:** Choose this if you're using Django REST Framework and want AI coding agents to help with frontend integration by accessing your API documentation. **When NOT to choose:** Don't choose this if you need the AI to directly call your API endpoints (this server only exposes documentation, not actions). ### [mcp-config-manager](https://mcpcatalogs.com/en/server/holstein13-mcp-config-manager) by holstein13 · score 43 · ★28 Cross-platform MCP server configuration manager with GUI for managing servers across Claude, Gemini & other AI systems. **When to choose:** Choose this tool if you manage MCP servers across multiple AI clients and want a visual interface for configuration management without editing JSON files manually. **When NOT to choose:** Don't choose this if you only use a single AI client with MCP servers and prefer direct configuration file editing. ### [mcp-rb](https://mcpcatalogs.com/en/server/funwarioisii-mcp-rb) by funwarioisii · score 43 · ★205 Ruby framework for building MCP servers with a Sinatra-like DSL, supporting resources and tools. **When to choose:** Choose MCP-RB if you're building Ruby-based MCP servers and prefer a simple, Sinatra-like DSL approach with minimal setup. **When NOT to choose:** Avoid MCP-RB if you need advanced MCP capabilities not yet implemented or if you prefer frameworks in other languages like Python or JavaScript. ### [Kiln](https://mcpcatalogs.com/en/server/codeofaxel-kiln) by codeofaxel · score 43 · ★18 Kiln is an MCP server that lets AI agents control 3D printers to design, generate, slice, and print physical objects. **When to choose:** Choose Kiln when you need AI agents to fully control 3D printing workflows, especially with multiple printer types or when you want version control for 3D designs. **When NOT to choose:** Don't choose Kiln if you only need basic printer control without AI integration, or if you're looking for free open-source solution as Kiln's advanced features require paid tiers. ### [webmcp-react](https://mcpcatalogs.com/en/server/mcpcat-webmcp-react) by MCPCat · score 43 · ★24 React hooks for exposing app functionality as WebMCP tools with type safety and SSR support. **When to choose:** Choose webmcp-react when building React applications that need to expose web-based functionality to AI agents with proper TypeScript typing. **When NOT to choose:** Don't choose if you need server-side tools or are building for non-React environments, as this is specifically a React implementation of WebMCP. ### [ue5-mcp](https://mcpcatalogs.com/en/server/mirno-ehf-ue5-mcp) by mirno-ehf · score 43 · ★37 UE5 MCP server plugin enabling AI agents to edit Unreal Engine Blueprints through natural language commands. **When to choose:** Choose UE5 MCP if you're working with Unreal Engine Blueprints and want to leverage AI assistants for development tasks through natural language commands. **When NOT to choose:** Avoid this if you need AI integration with non-Unreal Engine workflows or if you require authentication/permission controls for your development environment. ### [shebe](https://mcpcatalogs.com/en/server/shebe-oss-shebe) by shebe-oss · score 43 · ★32 Fast BM25 code search tool with 14 MCP tools for AI coding agents to find references and search codebases. **When to choose:** Choose Shebe when working with large codebases (1k+ files) where you need fast, ranked code search without cloud dependencies or GPU resources. **When NOT to choose:** Don't choose Shebe if you need semantic search capabilities, remote repository search, or if you're working exclusively with small codebases where simpler tools like ripgrep suffice. ### [ARBuilder](https://mcpcatalogs.com/en/server/quantum3-labs-arbuilder) by Quantum3-Labs · score 43 · ★21 AI-powered development assistant for Arbitrum ecosystem with 19 MCP tools providing code generation for Stylus, SDKs, dApps and Orbit chains. **When to choose:** Choose ARBuilder when developing applications for the Arbitrum ecosystem, especially if you need rapid prototyping of smart contracts, cross-chain bridges, or full-stack dApps with minimal boilerplate code. **When NOT to choose:** Avoid ARBuilder if you're working outside the Arbitrum ecosystem, need offline capabilities without API access, or require more granular control over the code generation process without relying on external services. ### [cucumberstudio-mcp](https://mcpcatalogs.com/en/server/herosizy-cucumberstudio-mcp) by HeroSizy · score 43 · ★21 MCP Server for Cucumber Studio providing access to test scenarios, action words, and test runs from the testing platform. **When to choose:** Choose this MCP server if you're using Cucumber Studio for test management and want AI assistants to access your test scenarios, action words, and test execution results. **When NOT to choose:** Avoid if you need write access to Cucumber Studio (this server is read-only only) or if you're not already invested in the Cucumber Studio ecosystem. ### [railway-mcp](https://mcpcatalogs.com/en/server/jason-tan-swe-railway-mcp) by jason-tan-swe · score 43 · ★72 A TypeScript MCP server for Railway.app that enables infrastructure management through natural language commands. **When to choose:** Choose railway-mcp if you use Railway.app for deployments and want to manage it through AI assistants like Claude, especially when combined with terminal-capable clients like Cursor for full deployment automation. **When NOT to choose:** Avoid this if you need full database template support, automatic GitHub repository linking, or are looking for a mature project with all features implemented (many are still marked as under development). ### [skill-to-mcp](https://mcpcatalogs.com/en/server/biocontext-ai-skill-to-mcp) by biocontext-ai · score 43 · ★24 Converts Claude Skills to MCP server resources with discovery and access tools. **When to choose:** Choose this when you have existing Claude Skills collections and want to expose them through MCP without rewriting them. **When NOT to choose:** Avoid this if you need to create skills from scratch, as it requires skills to be in the Claude Skills format rather than other formats. ### [open-code-review](https://mcpcatalogs.com/en/server/raye-deng-open-code-review) by raye-deng · score 43 · ★23 AI-powered code quality gate that detects hallucinated packages, stale APIs, and security anti-patterns in AI-generated code. **When to choose:** Choose this MCP server if you need to detect AI-specific code issues like hallucinated packages or stale APIs in AI-generated code, especially when working with multiple AI coding assistants. **When NOT to choose:** Avoid this if you need a fully open-source license (uses BSL-1.1 which converts to Apache 2.0 only in 2030) or if you're primarily reviewing non-AI-generated code. ### [kwin-mcp](https://mcpcatalogs.com/en/server/isac322-kwin-mcp) by isac322 · score 43 · ★23 MCP server for KDE Plasma 6 Wayland GUI automation with 30 tools for isolated virtual sessions and live desktop interaction. **When to choose:** Choose kwin-mcp when you need to automate KDE Plasma 6 applications on Wayland with full input capabilities and both isolated testing and live desktop modes. **When NOT to choose:** Avoid kwin-mcp if you need to automate applications on non-KDE desktop environments or X11 systems, as it only supports KDE Plasma 6 on Wayland. ### [freecad-mcp](https://mcpcatalogs.com/en/server/contextform-freecad-mcp) by contextform · score 43 · ★73 FreeCAD MCP enables AI-powered 3D modeling and CAD automation through natural language commands in Claude **When to choose:** Choose FreeCAD MCP when you want to control FreeCAD through natural language commands and need to automate CAD workflows without learning complex GUI operations. **When NOT to choose:** Don't choose this if you need advanced FreeCAD scripting capabilities not exposed through MCP, or if you're already invested in other CAD ecosystems with AI integration. ### [appstore-connect-mcp](https://mcpcatalogs.com/en/server/trialanderrorai-appstore-connect-mcp) by TrialAndErrorAI · score 43 · ★20 A code mode MCP server for Apple App Store Connect with 923 endpoints accessible through just 2 tools. **When to choose:** Choose this for development work with App Store Connect where you need programmatic access to the API without the overhead of maintaining 923 individual tool definitions. **When NOT to choose:** Avoid if you need a simpler tool-based interface without JavaScript coding, or if you require access to endpoints not covered by Apple's official API. ### [spicebridge](https://mcpcatalogs.com/en/server/clanker-lover-spicebridge) by clanker-lover · score 43 · ★19 MCP server providing AI assistants with direct access to ngspice circuit simulation. **When to choose:** Choose SPICEBridge when you need AI assistance for electronic circuit design and simulation, particularly if you want to work with SPICE simulations through natural language. **When NOT to choose:** Don't choose this if you need specialized SPICE variants other than ngspice, or if you require integrated PCB design capabilities beyond schematic generation. ### [src-to-kb](https://mcpcatalogs.com/en/server/vezlo-src-to-kb) by vezlo · score 43 · ★32 MCP server that converts source code to searchable knowledge base with IDE integration. **When to choose:** Choose this MCP server when you need AI-powered search across large codebases and want direct IDE integration with Claude Code or Cursor. **When NOT to choose:** Avoid this if you need real-time code analysis or if you prefer an open-source license as it's AGPL v3 licensed. ### [mcp-chat-studio](https://mcpcatalogs.com/en/server/joecastrom-mcp-chat-studio) by JoeCastrom · score 43 · ★17 A powerful MCP testing tool with multi-provider LLM support and modern UI for testing and debugging MCP servers. **When to choose:** Choose MCP Chat Studio when you need comprehensive testing, debugging, and development tools for MCP servers with support for multiple LLM providers and advanced workflow management. **When NOT to choose:** Don't choose this if you need simple testing with minimal setup, as it has a learning curve and is designed for complex MCP development rather than basic tool usage. ### [it-tools-mcp](https://mcpcatalogs.com/en/server/wrenchpilot-it-tools-mcp) by wrenchpilot · score 43 · ★21 MCP server providing access to 121+ IT tools including encoding, hashing, networking, and text processing utilities. **When to choose:** Choose this server when you need broad access to IT utilities within an MCP-compatible environment and want to integrate common developer and sysadmin tasks into automated workflows. **When NOT to choose:** Avoid this if you need highly specialized tools not covered by the 121 available utilities, or if you require authentication/permission features beyond basic logging. ### [awesome-agent-skills-mcp](https://mcpcatalogs.com/en/server/shadowrootdev-awesome-agent-skills-mcp) by shadowrootdev · score 43 · ★23 MCP server providing 100+ AI agent skills from top organizations including Anthropic, Vercel, and Hugging Face. **When to choose:** Choose this server when you want to quickly enhance your AI agents with a wide variety of pre-built skills from industry leaders without implementing each one manually. **When NOT to choose:** Don't choose this server if you need highly specialized skills not included in the curated collection, or if you need tight integration with custom authentication systems. ### [unified-gateway-mcp](https://mcpcatalogs.com/en/server/mcp360-unified-gateway-mcp) by mcp360 · score 43 · ★20 MCP360 is a unified gateway connecting AI agents to 100+ tools through a simple search and execute interface. **When to choose:** Choose MCP360 when you need access to multiple specialized tools without managing individual MCP configurations, especially for SEO, email verification, and market research workflows. **When NOT to choose:** Avoid if you need direct control over individual MCP server configurations, have strict data residency requirements, or prefer open-source solutions without vendor lock-in. ### [remembrallmcp](https://mcpcatalogs.com/en/server/roboticforce-remembrallmcp) by roboticforce · score 43 · ★18 MCP server providing persistent memory and code dependency graph for AI agents using Rust and Postgres. **When to choose:** Teams already using PostgreSQL who need persistent memory for AI agents and want to reduce token consumption in code analysis tasks. **When NOT to choose:** If you're not already using PostgreSQL or need write access to the database, as RemembrallMCP requires a Postgres + pgvector setup and primarily provides read-only access. ### [svelte5-mcp](https://mcpcatalogs.com/en/server/studentofjs-svelte5-mcp) by StudentOfJS · score 43 · ★64 MCP server providing specialized tools and knowledge for Svelte 5 development with runes, snippets, and reactivity patterns. **When to choose:** Choose this server when working specifically with Svelte 5 development and needing quick access to knowledge about runes, best practices, and code examples. **When NOT to choose:** Avoid if you need support for Svelte 4 or earlier versions, as this server focuses exclusively on Svelte 5 features and patterns. ### [BugPack](https://mcpcatalogs.com/en/server/duhuazhu-bugpack) by duhuazhu · score 43 · ★22 BugPack packages bug screenshots into AI-ready fix instructions with an MCP server. **When to choose:** Choose BugPack if you're an individual developer or team that frequently reports and fixes bugs from screenshots, and want AI coding agents to directly access and fix these issues without manual copy-paste. **When NOT to choose:** Don't choose BugPack if you need a cloud-based bug tracking system with multi-user collaboration features, as it's designed for local-only use with no built-in sharing capabilities. ### [astro-mcp](https://mcpcatalogs.com/en/server/morinokami-astro-mcp) by morinokami · score 43 · ★58 MCP server providing Astro project insights, runtime info, documentation search, and integration data. **When to choose:** Choose this if you're working with Astro projects and want AI assistants to have better understanding of your project structure, configurations, and access to up-to-date documentation. **When NOT to choose:** Avoid using this in production as it's explicitly marked as experimental and unstable. Consider alternatives if you need a more mature solution. ### [mcpx](https://mcpcatalogs.com/en/server/aigc-hackers-mcpx) by AIGC-Hackers · score 43 · ★20 A command-line tool that exposes remote MCP servers as agent-friendly commands with OAuth support and schema discovery. **When to choose:** Choose mcpx when building agent systems that need consistent interfaces to multiple MCP services without managing configuration files per project. **When NOT to choose:** Avoid mcpx if you need fine-grained control over tool implementations, require custom authentication flows beyond OAuth, or work exclusively with Windows environments. ### [cmcp](https://mcpcatalogs.com/en/server/russellluo-cmcp) by RussellLuo · score 43 · ★19 cMCP is a command-line utility for interacting with MCP servers over STDIO, HTTP, or SSE transports. **When to choose:** Choose cMCP when you need a simple, command-line interface to interact with MCP servers, especially for testing, debugging, or scripting purposes. **When NOT to choose:** Don't choose cMCP if you need a full GUI client with advanced features like conversation history or visual resource management. ### [hamr](https://mcpcatalogs.com/en/server/akhilraghav0-hamr) by AKhilRaghav0 · score 43 · ★18 A Go framework for building MCP servers with automatic schema generation from struct tags, optimizing for token cost. **When to choose:** Teams building MCP servers in Go who want to minimize boilerplate and optimize token costs, especially when working with multiple tools. **When NOT to choose:** If you need the latest MCP protocol features, official backing, or output schemas/progress tracking which aren't yet supported in hamr. ### [Unity-AI-Tools-Template](https://mcpcatalogs.com/en/server/ivanmurzak-unity-ai-tools-template) by IvanMurzak · score 43 · ★18 Template for creating Unity MCP tools with initialization scripts and setup guidance. **When to choose:** Choose this template when developing Unity-specific MCP tools and you need a structured approach to package creation, distribution, and maintenance. **When NOT to choose:** Don't choose this if you're not working with Unity Engine, as the template is specifically designed for Unity MCP tool development. ### [tesseron](https://mcpcatalogs.com/en/server/brainblend-ai-tesseron) by BrainBlend-AI · score 43 · ★17 Tesseron is a TypeScript SDK that exposes typed web-app actions to MCP-compatible AI agents over WebSocket. **When to choose:** Choose Tesseron when you need to expose real application functionality to AI agents through typed actions rather than automation, and when you're working with multiple client environments and want a consistent experience across different AI agents. **When NOT to choose:** Avoid Tesseron if you need extensive modification of the original code due to its BUSL-1.1 license restriction, or if you're looking for a simpler solution that doesn't require setting up a WebSocket gateway. ### [onetool-mcp](https://mcpcatalogs.com/en/server/beycom-onetool-mcp) by beycom · score 43 · ★18 OneTool MCP provides 100+ tools in a single server, reducing token usage by 96% and eliminating context rot. **When to choose:** Choose OneTool when you need access to 100+ developer tools and want to minimize token consumption, especially if you're already using Python for agent development. **When NOT to choose:** Don't choose OneTool if you prefer declarative tool definitions over code execution, or if you need tight integration with non-Python environments. ### [yamcp](https://mcpcatalogs.com/en/server/hamidra-yamcp) by hamidra · score 43 · ★66 YAMCP is a workspace manager that organizes and exposes multiple MCP servers through a unified gateway for AI applications. **When to choose:** Choose YAMCP when you need to organize multiple MCP servers into logical groups and manage them through a unified interface with centralized logging. **When NOT to choose:** Don't choose YAMCP if you're only using a single MCP server or prefer direct integration without the additional layer of abstraction. ### [continuum](https://mcpcatalogs.com/en/server/pouyahasanamreji-continuum) by pouyahasanamreji · score 43 · ★24 Shared memory + orchestration for coding agents via one MCP server with persistent vector memory. **When to choose:** Choose Continuum when you need multiple AI agents to collaborate on a long-term project while maintaining consistent knowledge and coordination. **When NOT to choose:** Don't choose Continuum if you're working on short-term, isolated tasks or prefer cloud-based solutions over local-first architectures. ### [bilig](https://mcpcatalogs.com/en/server/proompteng-bilig) by proompteng · score 43 · ★23 Formula workbook runtime exposing spreadsheet tools via MCP for Node.js services and agents. **When to choose:** Choose bilig when you need spreadsheet-like calculations in Node.js services or agent tools that require formula verification and persistence rather than visual editing. **When NOT to choose:** Don't choose bilig if you need visual spreadsheet editing, Office macro compatibility, or desktop Excel automation - it's designed for headless, programmatic use cases only. ### [apifox-mcp](https://mcpcatalogs.com/en/server/iwen-conf-apifox-mcp) by iwen-conf · score 43 · ★33 MCP server enabling AI assistants to manage Apifox API projects through natural language commands. **When to choose:** Choose this MCP server if you're already using Apifox for API management and want to integrate AI assistants into your workflow for documentation and API creation. **When NOT to choose:** Avoid this if you need API management functionality outside of Apifox, or if you're concerned about vendor lock-in to a specific API documentation platform. ### [mcp-gateway](https://mcpcatalogs.com/en/server/unrelated-ai-mcp-gateway) by unrelated-ai · score 43 · ★19 A gateway solution that transforms any HTTP endpoint into MCP servers, offering multi-tenant isolation and centralized management. **When to choose:** Choose this when you need to expose multiple existing systems as MCP servers with proper multi-tenant isolation and access controls, especially in enterprise environments where security and policy enforcement are critical. **When NOT to choose:** Don't choose this for simple, single-tenant MCP use cases where the overhead of a full gateway deployment isn't justified, or when you need direct tool access without additional proxy layers. ### [weppy-roblox-mcp](https://mcpcatalogs.com/en/server/hope1026-weppy-roblox-mcp) by hope1026 · score 43 · ★21 An MCP server that connects AI agents to Roblox Studio, enabling direct manipulation of game elements through natural language. **When to choose:** Choose WEPPY if you're developing Roblox games and want to use AI agents directly within Studio for prototyping, bulk editing, or automated testing of your game components. **When NOT to choose:** Don't choose WEPPY if you need access to Roblox Studio on platforms other than Windows 10+ or macOS 12+, or if you're unwilling to use the AGPL-3.0 licensed version or pay for the Pro upgrade. ### [code-to-tree](https://mcpcatalogs.com/en/server/micl2e2-code-to-tree) by micl2e2 · score 43 · ★83 A runtime-free MCP server converting source code to AST using tree-sitter with minimal dependencies. **When to choose:** Choose this server when you need AST conversion for source code analysis across multiple languages without adding runtime dependencies to your environment. **When NOT to choose:** Don't choose this if you need AST generation for languages not supported by tree-sitter, or if you require server-side processing capabilities rather than client-side conversion. ### [matlab-mcp-tools](https://mcpcatalogs.com/en/server/neuromechanist-matlab-mcp-tools) by neuromechanist · score 43 · ★18 MCP server for MATLAB development with script execution, workspace management and figure analysis tools. **When to choose:** Choose this MCP server if you work with MATLAB regularly and want to integrate it into AI coding assistants like Claude Code or Cursor, particularly if you need to execute scripts section by section or analyze MATLAB figures programmatically. **When NOT to choose:** Avoid this if you don't have a licensed MATLAB installation or need to execute code in a restricted environment (MATLAB Python engine has specific requirements and dependencies). ### [ai-mate](https://mcpcatalogs.com/en/server/symfony-ai-mate) by symfony · score 43 · ★22 MCP server for Symfony PHP projects that provides AI development assistance through standardized tools. **When to choose:** Developers working with PHP and Symfony projects who want AI assistance directly integrated into their development workflow. **When NOT to choose:** You're not working with PHP/Symfony projects, or you need production-ready functionality rather than development assistance. ### [pyxel-mcp](https://mcpcatalogs.com/en/server/kitao-pyxel-mcp) by kitao · score 43 · ★17 MCP server for AI-assisted retro game development with Pyxel, offering tools to run, capture, and analyze games. **When to choose:** Choose this MCP server if you're working with retro game development using Pyxel and want AI assistance with automated testing, visual analysis, and code validation. **When NOT to choose:** Don't choose this if you're working with modern game engines like Unity or Unreal Engine, as it's specifically designed for Pyxel's retro game development workflow. ### [codeprism](https://mcpcatalogs.com/en/server/rustic-ai-codeprism) by rustic-ai · score 43 · ★27 100% AI-generated MCP server providing code intelligence through graph-based analysis of codebases. **When to choose:** Choose CodePrism when you need comprehensive code intelligence with graph-based understanding of codebases, especially for JavaScript/TypeScript and Python projects. **When NOT to choose:** Avoid using CodePrism if you need human-reviewed code quality or require support for programming languages beyond JavaScript/TypeScript and Python. ### [better-godot-mcp](https://mcpcatalogs.com/en/server/n24q02m-better-godot-mcp) by n24q02m · score 43 · ★18 Composite MCP server providing 17 mega-tools for AI-assisted Godot game development. **When to choose:** Choose this MCP server for AI-assisted Godot game development with comprehensive tooling for scenes, nodes, scripts, shaders, and game engine features. **When NOT to choose:** Don't choose if you need Godot 3.x compatibility or require write access to internal Godot APIs (this server works with text file manipulation only). ### [llm-app-exploration](https://mcpcatalogs.com/en/server/forrestkim42-llm-app-exploration) by ForrestKim42 · score 43 · ★23 MCP servers for automated app exploration via accessibility APIs, building complete UI maps and transition tables. **When to choose:** Choose this when you need systematic, comprehensive app exploration rather than vision-based UI automation, especially for accessibility-first analysis. **When NOT to choose:** Don't choose this for vision-based UI automation or when you need cross-platform support beyond the specifically implemented platforms (Android, iOS, macOS). ### [api200](https://mcpcatalogs.com/en/server/api-200-api200) by API-200 · score 43 · ★255 API200 is an API gateway with MCP server integration for managing third-party APIs with authentication, caching, and monitoring. **When to choose:** Choose API 200 when you need an API gateway for third-party integrations and also want to expose those APIs through MCP. **When NOT to choose:** Don't choose API 200 if you only need MCP server functionality without the gateway features, as it adds unnecessary complexity. ### [hoot](https://mcpcatalogs.com/en/server/portkey-ai-hoot) by Portkey-AI · score 43 · ★19 Hoot is a testing tool for MCP servers that provides a Postman-like interface for testing tools, authentication, and exploring MCP servers. **When to choose:** Choose Hoot when you need a visual tool for testing MCP servers with proper OAuth support and session persistence, especially for complex authentication flows. **When NOT to choose:** Don't choose Hoot if you need direct MCP server implementation rather than a testing tool, or if you require support for MCP resources and prompts which are still in development. ### [primitiv](https://mcpcatalogs.com/en/server/ai-by-design-primitiv) by AI-by-design · score 42 · ★16 Primitiv creates a single design contract from multiple sources via MCP, ensuring UI agents build consistently. **When to choose:** Choose Primitiv when your team uses multiple design tools and needs AI agents to build UI that strictly adheres to a reconciled design system contract. **When NOT to choose:** Avoid Primitiv if you're working with a single design source and don't need reconciliation, or if you prefer human-readable documentation over machine-readable contracts. ### [domain-search-mcp](https://mcpcatalogs.com/en/server/dorukardahan-domain-search-mcp) by dorukardahan · score 42 · ★20 A zero-config domain search MCP server with multi-TLD checks, bulk search, premium detection, AI-powered suggestions, and registrar pricing comparison. **When to choose:** Choose this server when you need to check domain availability without API keys or want AI-powered domain suggestions integrated into your workflow. **When NOT to choose:** Don't choose this if you need real-time pricing without setting up a custom backend, or if you require domain registration functionality directly from the tool. ### [mcp-salesforce](https://mcpcatalogs.com/en/server/aiondadotcom-mcp-salesforce) by AiondaDotCom · score 42 · ★16 Complete MCP server for Salesforce integration with Claude Desktop, offering OAuth authentication and universal CRUD operations. **When to choose:** Choose this server when you need a complete, out-of-the-box integration between AI assistants and Salesforce with minimal setup requirements. **When NOT to choose:** Avoid if you're not using macOS (required for secure credential storage) or if you need advanced Salesforce features not covered by the universal CRUD interface. ### [eraser-io-mcp-server](https://mcpcatalogs.com/en/server/buck-0x-eraser-io-mcp-server) by buck-0x · score 42 · ★21 A Python MCP server for rendering various diagram types using the Eraser API with customizable themes and output formats. **When to choose:** Choose this MCP server when you need to generate diagrams programmatically within AI environments and are already using or willing to use Eraser as your diagramming tool. **When NOT to choose:** Avoid this if you need diagramming capabilities with a different provider, as this is tightly integrated with Eraser and requires an API token from their service. ### [robotmcp_client](https://mcpcatalogs.com/en/server/robotmcp-robotmcp-client) by robotmcp · score 42 · ★12 MCP client for ROS robot integration with Gemini, Ollama and other LLMs. **When to choose:** Choose this when you need to connect LLMs with ROS/ROS2 robots via MCP protocol for natural language interaction, testing, or autonomous control. **When NOT to choose:** Not suitable if you don't have ROS or ROS2 environments set up, as it requires both ros-mcp-server and rosbridge to function properly. ### [ui-ux-pro-mcp](https://mcpcatalogs.com/en/server/redf0x1-ui-ux-pro-mcp) by redf0x1 · score 42 · ★17 AI-powered UI/UX design intelligence MCP server with 1,920+ curated design resources and search tools. **When to choose:** Choose this server when you need specialized UI/UX design resources and are working with React, Vue, Flutter, or cross-platform applications. **When NOT to choose:** Don't choose this if you need comprehensive design tools beyond search capabilities or require real-time design collaboration features. ### [memory-journal-mcp](https://mcpcatalogs.com/en/server/neverinfamous-memory-journal-mcp) by neverinfamous · score 42 · ★16 An MCP server providing AI memory with project detection, GitHub integration, and knowledge graphs for context-aware development. **When to choose:** Choose this MCP server if you're working on GitHub projects that need persistent memory between AI sessions and require deep GitHub integration for workflow automation and team collaboration. **When NOT to choose:** Avoid if you need a lightweight solution or are not working with GitHub, as this server is feature-rich but may be overkill for simpler use cases or projects not using GitHub. ### [nestjs-mcp](https://mcpcatalogs.com/en/server/orbit-codes-nestjs-mcp) by orbit-codes · score 42 · ★11 A NestJS integration for MCP that enables building MCP servers using dependency injection and decorators. **When to choose:** Choose this if you're already using NestJS and want to build MCP servers while leveraging dependency injection and decorator patterns. **When NOT to choose:** Avoid if you're not using NestJS or prefer a more lightweight MCP implementation without framework dependencies. ### [mcp-domain-availability](https://mcpcatalogs.com/en/server/imprvhub-mcp-domain-availability) by imprvhub · score 42 · ★49 A domain availability MCP server that checks domain names across 50+ TLDs with DNS/WHOIS verification. **When to choose:** Choose this server if you need to check domain availability for multiple TLDs within Claude Desktop and want smart suggestions based on availability. **When NOT to choose:** Don't choose this if you need domain registration capabilities (this is read-only only) or require checking specialized/obscure TLDs not in its supported list. ### [reactbits-mcp-server](https://mcpcatalogs.com/en/server/ceorkm-reactbits-mcp-server) by ceorkm · score 42 · ★46 MCP server providing AI assistants with access to 135+ animated React components from ReactBits.dev. **When to choose:** Choose this server when you're working on React projects and need AI assistance with finding and implementing animated components from ReactBits.dev. **When NOT to choose:** Don't choose this if you need button, form, or loader components as they're incomplete, or if you're not working with React/JavaScript projects. ### [LSP4J-MCP](https://mcpcatalogs.com/en/server/stephanj-lsp4j-mcp) by stephanj · score 42 · ★23 Java MCP server wrapping JDTLS to provide Java IDE features like symbol search and reference finding to AI assistants. **When to choose:** Choose this when you need AI assistants to understand and work with Java codebases using familiar IDE-like functionality without requiring complex setup. **When NOT to choose:** Don't choose this if you're working with non-Java languages or need advanced refactoring features not covered by basic LSP operations. ### [GUI-MCP](https://mcpcatalogs.com/en/server/phialsbasement-gui-mcp) by PhialsBasement · score 42 · ★24 Visual node editor for creating FastMCP servers without coding, featuring Blueprint-style interface. **When to choose:** Choose GUI-MCP when you need to create MCP servers without programming knowledge or prefer visual programming over writing code. **When NOT to choose:** Avoid GUI-MCP if you need fine-grained control over your MCP server implementation or prefer working directly with code rather than a visual interface. ### [mcp-framework](https://mcpcatalogs.com/en/server/koki7o-mcp-framework) by koki7o · score 42 · ★22 Rust framework for building AI agents with native MCP support, multi-LLM integration, and web inspector. **When to choose:** Choose this framework for production AI agents requiring high performance, memory safety, and the ability to connect to multiple MCP servers simultaneously. **When NOT to choose:** Not ideal if you need rapid prototyping in languages other than Rust or require features beyond the current roadmap (resources and prompts protocols are planned but not yet implemented). ### [mcp-diagnostics-extension](https://mcpcatalogs.com/en/server/newbpydev-mcp-diagnostics-extension) by newbpydev · score 42 · ★22 VS Code extension that exposes diagnostic problems via MCP for AI agents to access in real-time. **When to choose:** Choose this if you're using VS Code and want your AI assistants to have real-time visibility into your code diagnostics without manual intervention. **When NOT to choose:** Don't choose this if you need diagnostic access from IDEs other than VS Code, or if you require write capabilities to diagnostic information (this is read-only). ### [toolkit-mcp-server](https://mcpcatalogs.com/en/server/cyanheads-toolkit-mcp-server) by cyanheads · score 42 · ★18 TypeScript-based MCP server providing system utilities, network diagnostics, security tools, and QR code generation. **When to choose:** Choose this server when you need system utilities, network diagnostics, and security tools integrated into your LLM workflow, especially if you're already using TypeScript-based applications. **When NOT to choose:** Don't choose this if you need high-volume usage beyond 45 requests/minute, or if you require database-specific operations rather than system utilities. ### [mcp-streamable-http](https://mcpcatalogs.com/en/server/invariantlabs-ai-mcp-streamable-http) by invariantlabs-ai · score 42 · ★129 Python and TypeScript MCP server examples implementing Streamable HTTP protocol for cross-language communication. **When to choose:** Choose this reference implementation when you need to build MCP servers or clients using the Streamable HTTP transport protocol, especially when working with cross-language implementations. **When NOT to choose:** Don't choose this if you need a production-ready solution with robust error handling, security features, or authentication - this is primarily an example implementation. ### [Genexus18MCP](https://mcpcatalogs.com/en/server/lennix1337-genexus18mcp) by lennix1337 · score 42 · ★16 GeneXus 18 MCP server enabling AI agents to read, edit, and analyze KB objects via Model Context Protocol. **When to choose:** Choose this if you work with GeneXus applications and want AI assistance to directly manipulate knowledge base objects through natural language. **When NOT to choose:** Don't choose this if you need cross-platform support (GeneXus is Windows-only) or if you're using a version of GeneXus other than 18. ### [rbxdev-ls](https://mcpcatalogs.com/en/server/0neshot101-rbxdev-ls) by 0neShot101 · score 42 · ★16 Roblox-focused MCP server that provides tools for interacting with live game instances and code execution. **When to choose:** Choose this MCP server if you're developing Roblox games and want AI assistants to be able to inspect, analyze, and manipulate game state in real-time. **When NOT to choose:** Don't choose this if you're not working with Roblox games or if you need a more general-purpose game development MCP server that supports multiple game engines. ### [cli-use](https://mcpcatalogs.com/en/server/justvugg-cli-use) by JustVugg · score 42 · ★16 A tool that converts any MCP server into a CLI, reducing token costs by 60-80% with zero runtime dependencies. **When to choose:** Choose cli-use when you need to reduce token costs for MCP tool calls or want to use MCP servers as native Unix-style CLI commands in shell scripts. **When NOT to choose:** Don't choose cli-use if you need full MCP protocol features beyond basic tool calling or require real-time response times below 10ms. ### [swift-skeleton](https://mcpcatalogs.com/en/server/1amageek-swift-skeleton) by 1amageek · score 42 · ★15 Swift code skeleton extractor that provides structural overviews of codebases via MCP for better LLM code exploration. **When to choose:** Choose swift-skeleton for large Swift or multi-language codebases where you need AI agents to understand structure without exceeding context limits. **When NOT to choose:** Don't choose it if you need write access to codebases or support for languages not in its parser list. ### [opentrace](https://mcpcatalogs.com/en/server/adham90-opentrace) by adham90 · score 42 · ★15 OpenTrace is an MCP-native observability engine that connects AI coding agents directly to production data via MCP tools. **When to choose:** Choose OpenTrace if you're already using Postgres and want a low-cost, AI-native observability solution that directly connects your coding assistant to production data without UI dashboards. **When NOT to choose:** Don't choose OpenTrace if you need complex dashboards for human analysts, require write access to databases, or need multi-tenant capabilities with isolated data. ### [faf](https://mcpcatalogs.com/en/server/wolfe-jam-faf) by Wolfe-Jam · score 42 · ★15 FAF is an IANA-registered format that provides structured AI context through project.faf files and MCP servers. **When to choose:** Choose FAF when working across multiple AI coding assistants and need a standardized format to share project context consistently. **When NOT to choose:** Avoid FAF if you only use a single AI coding assistant and prefer to maintain separate instruction files for each tool. ### [taiga-ui-mcp](https://mcpcatalogs.com/en/server/taiga-family-taiga-ui-mcp) by taiga-family · score 42 · ★13 MCP server providing Taiga UI documentation, component examples, and migration guides for Angular development. **When to choose:** Choose this MCP server if you're working with Angular and Taiga UI components and need AI-assisted access to documentation, code examples, and migration guides within your development workflow. **When NOT to choose:** Avoid this if you're not using Angular or Taiga UI, as it provides specialized functionality specific to this UI library ecosystem. ### [clawmaster](https://mcpcatalogs.com/en/server/openmaster-ai-clawmaster) by openmaster-ai · score 42 · ★20 A comprehensive MCP server platform for managing OpenClaw with web console, desktop app, and MCP server integration. **When to choose:** Choose ClawMaster when you need a graphical interface to manage OpenClaw instances, especially if you're non-technical or managing for a team. **When NOT to choose:** Don't choose ClawMaster if you're looking for MCP server implementations or if you prefer CLI-based configuration. ### [devcontext](https://mcpcatalogs.com/en/server/aiurda-devcontext) by aiurda · score 42 · ★45 Autonomous MCP server providing project-centric context awareness through pattern learning and relationship graphs. **When to choose:** Choose DevContext when working on large codebases where maintaining context across development sessions is critical, and when you prefer non-vector retrieval methods over traditional embeddings. **When NOT to choose:** Don't choose DevContext if you need vector-based semantic search, if you're already invested in other context systems, or if you prefer solutions with more language support beyond the six officially supported languages. ### [MCP-Developer-SubAgent](https://mcpcatalogs.com/en/server/gensecaihq-mcp-developer-subagent) by gensecaihq · score 42 · ★28 Production-ready framework with 8 Claude Code sub-agents and FastMCP templates for MCP development. **When to choose:** Choose this framework when developing MCP servers with a focus on security, production readiness, and specialized AI assistance for both Claude Code integration and programmatic SDK usage. **When NOT to choose:** Don't choose this for simple MCP projects or if you prefer pure command-line tools without AI assistance, as it includes significant overhead from the AI orchestration components. ### [kai](https://mcpcatalogs.com/en/server/kaicontext-kai) by kaicontext · score 42 · ★16 Kai is a semantic analysis engine for code changes that provides MCP tools for call graphs, dependencies, and impact analysis. **When to choose:** Choose Kai when you need deep semantic understanding of code changes and dependencies, especially for CI optimization and AI-assisted code analysis across multiple programming languages. **When NOT to choose:** Don't choose Kai if you need write access to repositories (this is primarily a read-only semantic analysis tool) or if you're concerned about vendor lock-in to the kaicontext.com service. ### [mcp-time](https://mcpcatalogs.com/en/server/theobrigitte-mcp-time) by TheoBrigitte · score 42 · ★15 A Go-based MCP server providing time utilities with natural language processing and timezone conversions. **When to choose:** Choose this MCP server when your AI application needs robust time manipulation capabilities with natural language parsing, especially for applications that handle multiple timezones frequently. **When NOT to choose:** Don't choose this if you need complex calendar or event management features beyond simple time operations, as it focuses specifically on time utilities rather than full calendar functionality. ### [inspector](https://mcpcatalogs.com/en/server/mcp-use-inspector) by mcp-use · score 42 · ★13 A modern MCP Inspector for testing and debugging remote MCP servers with support for Apps SDK and browser-based UI. **When to choose:** Choose MCP Inspector when you need a comprehensive testing tool for MCP servers, especially for developers who want to interactively test tools, debug connections, and explore server capabilities before integration. **When NOT to choose:** Avoid if you need a production-ready MCP server implementation itself rather than a testing tool, or if you prefer minimal dependencies - the Inspector requires a Node.js environment and browser interface. ### [delimit-mcp-server](https://mcpcatalogs.com/en/server/delimit-ai-delimit-mcp-server) by delimit-ai · score 42 · ★16 MCP server for AI coding governance across multiple AI assistants with attestations and deploy gates. **When to choose:** Choose Delimit when you need to govern AI-assisted coding across multiple tools and require auditable attestations for all AI-generated code merges. **When NOT to choose:** Don't choose Delimit if you're working with a single AI coding tool and don't need audit trails or multi-model consensus features. ### [dap_mcp](https://mcpcatalogs.com/en/server/kashuncheng-dap-mcp) by KashunCheng · score 42 · ★40 DAP-MCP enables LLM interaction with debuggers through standardized MCP protocol, offering debugging tools like breakpoints and execution control. **When to choose:** Choose dap-mcp when you need AI assistance for debugging workflows and your team uses Python (debugpy) or lldb debuggers. **When NOT to choose:** Avoid if you use other debuggers not yet supported, or need a more comprehensive IDE integration beyond MCP protocol support. ### [mcp-gateway](https://mcpcatalogs.com/en/server/lightconetech-mcp-gateway) by lightconetech · score 42 · ★151 A gateway that bridges stdio-based Claude Desktop with HTTP/SSE MCP servers. **When to choose:** Choose this when you need to connect Claude Desktop to HTTP/SSE-based MCP servers but don't want to modify the server implementation. **When NOT to choose:** Avoid this if you only need stdio-based MCP servers or are looking for a more direct integration solution without an additional gateway layer. ### [dx-toolkit](https://mcpcatalogs.com/en/server/youdotcom-oss-dx-toolkit) by youdotcom-oss · score 42 · ★20 You.com DX Toolkit provides MCP server integration and SDK plugins for developers to leverage You.com AI capabilities. **When to choose:** Choose this when you need to integrate You.com's AI capabilities into your workflow and want standardized MCP protocol implementations for multiple AI frameworks. **When NOT to choose:** Avoid if you need to avoid vendor lock-in to You.com's services or require self-hosted MCP capabilities without third-party dependencies. ### [openapi-to-mcp](https://mcpcatalogs.com/en/server/evilfreelancer-openapi-to-mcp) by EvilFreelancer · score 42 · ★16 OpenAPI-to-MCP converts any OpenAPI/Swagger API into MCP tools with HTTP execution. **When to choose:** Choose this when you have an existing REST API with OpenAPI documentation and want to quickly enable AI clients to interact with it without backend modifications. **When NOT to choose:** Don't choose this if you need complex authentication flows beyond Basic Auth/Bearer tokens, or if your API requires significant customization for AI clients. ### [python-notebook-mcp](https://mcpcatalogs.com/en/server/usamak98-python-notebook-mcp) by UsamaK98 · score 42 · ★45 A Python MCP server for AI assistants to create, edit and view Jupyter notebooks. **When to choose:** Choose this server when working with Jupyter notebooks in AI environments that need programmatic access to notebook files without full execution capabilities. **When NOT to choose:** Don't choose this if you need actual code execution capabilities within the notebooks or if you're working with non-Jupyter notebook formats. ### [codebase-mcp](https://mcpcatalogs.com/en/server/danyqe-codebase-mcp) by danyQe · score 42 · ★43 Privacy-first AI coding assistant that turns Claude into a powerful MCP-based development tool with semantic search, memory, and quality-checked code generation. **When to choose:** Choose this if you already have a Claude subscription and want privacy-first AI coding assistance without paying additional monthly fees for tools like Cursor or Windsurf. **When NOT to choose:** Avoid this if you need edit operations beyond Gemini's free tier limits (15 RPM, 250K TPM) or if you require support for languages other than Python, JavaScript, TypeScript, and FastAPI. ### [mcp-server](https://mcpcatalogs.com/en/server/configcat-mcp-server) by configcat · score 42 · ★15 ConfigCat MCP server provides tool access to feature flag and configuration management through ConfigCat's public API. **When to choose:** Teams using ConfigCat for feature flag management who want to integrate flag operations directly into their development workflow via MCP. **When NOT to choose:** Avoid if you need runtime evaluation of feature flags (this server is for management only) or if you're not already using ConfigCat. ### [local-history-mcp](https://mcpcatalogs.com/en/server/xxczaki-local-history-mcp) by xxczaki · score 42 · ★15 MCP server providing AI access to VS Code/Cursor's Local History for data recovery. **When to choose:** Choose this if you're using VS Code or Cursor and need to recover lost code versions or compare file states without leaving your AI workflow. **When NOT to choose:** Don't use this if you need history recovery for other editors, or if you're working in an environment where Local History is disabled or not supported. ### [dynamic-fastmcp](https://mcpcatalogs.com/en/server/ragieai-dynamic-fastmcp) by ragieai · score 42 · ★45 Dynamic FastMCP extends MCP Python server with context-aware tools that adapt behavior based on user and tenant context. **When to choose:** Choose Dynamic FastMCP when building multi-tenant applications or tools that need to adapt their behavior based on user context, tenant information, or request parameters. **When NOT to choose:** Don't choose this if you need simple, static tools without context adaptation or if you're not building on Python with FastAPI. ### [payloadcmsmcp](https://mcpcatalogs.com/en/server/disruption-hub-payloadcmsmcp) by disruption-hub · score 41 · ★114 Specialized MCP server for Payload CMS 3.0 providing code validation, template generation, and project scaffolding tools. **When to choose:** Choose this if you're developing with Payload CMS 3.0 and need specialized tools for validation, template generation, and project scaffolding. **When NOT to choose:** Not suitable if you're not using Payload CMS, as it's specifically designed for this CMS platform and relies on Railway deployment. ### [sandbox-mcp](https://mcpcatalogs.com/en/server/pottekkat-sandbox-mcp) by pottekkat · score 41 · ★110 Sandbox MCP enables secure code execution in isolated Docker containers through the MCP protocol. **When to choose:** Choose Sandbox MCP when you need to safely execute code generated by LLMs in isolated environments, particularly for testing or educational purposes. **When NOT to choose:** Avoid Sandbox MCP if you need to execute code requiring persistent storage or network access beyond what the sandboxes provide, or if Docker is not available in your environment. ### [mcp-task-orchestrator](https://mcpcatalogs.com/en/server/echoingvesper-mcp-task-orchestrator) by EchoingVesper · score 41 · ★25 A sophisticated MCP server for AI-assisted development that provides task orchestration with specialized roles and persistent memory. **When to choose:** Choose this server for complex development projects that require structured workflows with specialist roles, persistent memory, and automated documentation throughout the development lifecycle. **When NOT to choose:** Avoid if you need simple, single-session task completion without persistent storage or specialized roles; the server is overkill for basic AI assistance needs. ### [pywss](https://mcpcatalogs.com/en/server/czasg-pywss) by czasg · score 41 · ★100 Pywss is a lightweight Python web framework with built-in MCP server capabilities supporting SSE, StreamHTTP and MCPO protocols. **When to choose:** Choose Pywss when you need a lightweight Python framework with built-in MCP server capabilities and want to avoid the overhead of WSGI-based frameworks. **When NOT to choose:** Don't choose Pywss if you need advanced ORM features, extensive built-in admin interfaces, or if your project already heavily invests in Flask/Django ecosystem. ### [mcp-jest](https://mcpcatalogs.com/en/server/reallyartificial-mcp-jest) by ReallyArtificial · score 41 · ★16 A testing framework for MCP servers, providing Jest-like syntax for automated testing, protocol validation, and CI/CD integration. **When to choose:** Choose MCP-Jest if you're developing MCP servers and need automated testing for your tools, resources, and prompts, especially if you're already using Jest or want CI/CD integration. **When NOT to choose:** Don't choose MCP-Jest if you only need simple validation or if your MCP server is already covered by comprehensive testing infrastructure. ### [modex](https://mcpcatalogs.com/en/server/theronic-modex) by theronic · score 41 · ★114 Modex is a Clojure library implementing MCP Server & Client to extend AI models with tools, resources, and prompts. **When to choose:** Choose Modex if you're working with Clojure and need a native implementation of MCP that doesn't require additional proxy layers. **When NOT to choose:** Don't choose Modex if you need resources or prompts support (currently WIP), or if you prefer the Apache 2.0 license over its GPL/commercial licensing model. ### [mcp-client-gen](https://mcpcatalogs.com/en/server/kriasoft-mcp-client-gen) by kriasoft · score 41 · ★16 Tool to generate type-safe TypeScript clients for any MCP server with OAuth 2.1 support. **When to choose:** Choose this when working with TypeScript and needing type-safe access to MCP servers, especially OAuth-protected ones. **When NOT to choose:** Don't choose if you need authentication methods beyond OAuth 2.1 or if you're working in a non-TypeScript environment. ### [mcpm](https://mcpcatalogs.com/en/server/mcp-club-mcpm) by MCP-Club · score 41 · ★108 A command-line tool for managing MCP servers in Claude App with package discovery and server control capabilities. **When to choose:** Choose MCPM when managing multiple MCP servers in Claude App and you need a unified CLI interface with package discovery capabilities. **When NOT to choose:** Don't choose MCPM if you primarily work with AI applications outside of Claude or need advanced server management features beyond its current scope. ### [mcp-server-playground](https://mcpcatalogs.com/en/server/chrisleekr-mcp-server-playground) by chrisleekr · score 41 · ★7 A TypeScript MCP server with HTTP transport, OAuth proxy support, and tools for AWS services, system operations, and data streaming. **When to choose:** Choose this when you need to build secure MCP services with OAuth integration, require reliable connection handling with SSE, or need to deploy MCP servers in clusters with stateful sessions. **When NOT to choose:** Avoid if you need a simple MCP server without OAuth complexity, are working in an environment without Node.js/Bun support, or require a fully managed service rather than a self-hosted solution. ### [mcp-openapi](https://mcpcatalogs.com/en/server/reapi-com-mcp-openapi) by ReAPI-com · score 41 · ★86 MCP server that loads and serves OpenAPI specifications to enable LLM-powered API integrations in IDEs. **When to choose:** Choose this server when working with multiple OpenAPI specs and need LLM integration in your IDE for API documentation, client generation, or schema understanding. **When NOT to choose:** Don't choose this if you need real-time API mocking or testing capabilities, as this server focuses on static API documentation and schema information. ### [mcp_server_gdb](https://mcpcatalogs.com/en/server/pansila-mcp-server-gdb) by pansila · score 41 · ★63 MCP server that exposes GDB debugging capabilities to AI assistants through standardized tools. **When to choose:** Choose this MCP server when you need program debugging capabilities through AI assistants, particularly for C/C++ applications, and prefer a native Rust implementation with concurrent session support. **When NOT to choose:** Avoid this if you need debugging capabilities for non-C/C++ languages, or if you require authentication mechanisms for secure multi-user environments, as none are currently documented. ### [stash-mcp](https://mcpcatalogs.com/en/server/diomonogatari-stash-mcp) by diomonogatari · score 41 · ★4 A production-ready MCP server for Atlassian Bitbucket Server with 40 tools and robust resilience features. **When to choose:** Choose this server if your team uses Bitbucket Server and wants AI assistants to have contextual access to repositories, PRs, and code reviews with workflow-optimized tools. **When NOT to choose:** Avoid if you use Bitbucket Cloud (this is for Server only), have strict security requirements preventing Docker deployment, or need functionality beyond what 40 specialized tools provide. ### [uipath-mcp-python](https://mcpcatalogs.com/en/server/uipath-uipath-mcp-python) by UiPath · score 41 · ★8 Python SDK for building and hosting MCP servers within UiPath automation platform. **When to choose:** Choose this SDK if you're working within the UiPath ecosystem and need to create MCP servers that integrate with UiPath workflows. **When NOT to choose:** Avoid this if you need MCP servers for platforms other than UiPath, as it creates vendor-specific dependencies. ### [fastify-mcp-server](https://mcpcatalogs.com/en/server/neddl-fastify-mcp-server) by NEDDL · score 41 · ★7 High-performance MCP server built with Fastify, TypeScript, and functional programming principles. **When to choose:** When you need a production-ready, high-performance MCP server with TypeScript support, session isolation, and enterprise features like authentication and metrics. **When NOT to choose:** If you need a simpler MCP implementation without session management or if you're not using Fastify/TypeScript in your stack. ### [sequential-thinking-mcp-v2](https://mcpcatalogs.com/en/server/spideynolove-sequential-thinking-mcp-v2) by spideynolove · score 41 · ★8 A structured reasoning MCP server with session persistence and memory management **When to choose:** Choose this for complex coding tasks requiring persistent memory across sessions and structured reasoning approaches. **When NOT to choose:** Avoid if you need immediate production stability, as this is undergoing architectural redesign with unclear implementation timeline. ### [microsoft_fabric_mcp](https://mcpcatalogs.com/en/server/augustab-microsoft-fabric-mcp) by Augustab · score 41 · ★25 MCP server providing read-only access to Microsoft Fabric resources via its REST API with 25 tools. **When to choose:** Choose this if you're working with Microsoft Fabric and need AI assistants to have read-only access to workspace structures, table schemas, and job information without data modification capabilities. **When NOT to choose:** Avoid this if you need write access to Microsoft Fabric resources or are not already using Microsoft Fabric in your workflow. ### [codemesh](https://mcpcatalogs.com/en/server/kiliman-codemesh) by kiliman · score 41 · ★25 CodeMesh is a self-improving MCP server that lets AI agents orchestrate multiple MCP servers through TypeScript code. **When to choose:** Choose CodeMesh when you need to orchestrate multiple MCP servers and want the system to self-improve through documentation of tool outputs. **When NOT to choose:** Don't choose CodeMesh if you need fine-grained control over individual tools or if your use case only requires a single MCP server. ### [mcp-ts](https://mcpcatalogs.com/en/server/zonlabs-mcp-ts) by zonlabs · score 41 · ★8 A comprehensive TypeScript SDK with Python bridge for implementing MCP servers and clients with multiple storage backends. **When to choose:** Choose mcp-ts when building MCP servers in TypeScript/JavaScript, especially with Next.js or serverless deployments, or when needing framework-specific adapters for popular AI development tools. **When NOT to choose:** Avoid if you need pure Python-based MCP implementation or if you require authentication mechanisms beyond the simple example shown in the documentation. ### [laravel-mcp-sdk](https://mcpcatalogs.com/en/server/mohamedahmed01-laravel-mcp-sdk) by mohamedahmed01 · score 41 · ★29 A comprehensive Laravel implementation of the Model Context Protocol with multiple transport options and tool registration. **When to choose:** Choose this server if you're building Laravel applications that need AI integration with Laravel's ecosystem features, require multiple transport options, or need to leverage existing Laravel infrastructure. **When NOT to choose:** Avoid this if you're working outside the Laravel ecosystem, need a pure PHP implementation without framework dependencies, or require very lightweight footprint. ### [nest-mcp](https://mcpcatalogs.com/en/server/btwld-nest-mcp) by btwld · score 41 · ★7 A NestJS-based framework for building MCP servers, clients and gateways with TypeScript decorators and dependency injection. **When to choose:** Choose nest-mcp if you're already using NestJS or want a TypeScript-based framework with decorators and dependency injection for building MCP servers. **When NOT to choose:** Don't choose this if you need a simple, lightweight solution without the NestJS framework overhead, or if you're building in JavaScript/TypeScript without the NestJS ecosystem. ### [any-script-mcp](https://mcpcatalogs.com/en/server/izumin5210-any-script-mcp) by izumin5210 · score 41 · ★35 MCP server exposing CLI tools and shell scripts as MCP tools through YAML configuration. **When to choose:** When you need to expose existing shell scripts or CLI tools as MCP tools without writing custom code, especially when working with complex AI agent integrations. **When NOT to choose:** If you need strict security boundaries for tool execution, as the server runs shell commands with the permissions of the user that started it. ### [mcp-ecosystem](https://mcpcatalogs.com/en/server/synkraai-mcp-ecosystem) by SynkraAI · score 41 · ★26 Collection of MCP server configurations, presets, and IDE configs for AIOS and Claude Code integration. **When to choose:** Choose this if you're using AIOS or Claude Code and want pre-configured MCP setups optimized for different development scenarios. **When NOT to choose:** Avoid if you need MCP servers not included in this collection, as this is a curated set rather than a comprehensive MCP server implementation. ### [godoc-mcp-server](https://mcpcatalogs.com/en/server/yikakia-godoc-mcp-server) by yikakia · score 41 · ★35 An MCP server that provides information about Go packages from pkg.go.dev. **When to choose:** Choose this MCP server when working with Go packages and need quick, AI-accessible documentation from pkg.go.dev without leaving your development environment. **When NOT to choose:** Don't choose this if you need documentation for languages other than Go or require more detailed package analysis beyond what pkg.go.dev provides. ### [unplugin-mcp](https://mcpcatalogs.com/en/server/situ2001-unplugin-mcp) by situ2001 · score 41 · ★31 A unified plugin for integrating MCP servers into JavaScript build tools like Webpack, Rollup, and Vite. **When to choose:** Choose this when you want AI assistants to understand and interact with your JavaScript build process and you're already using tools like Rollup or Webpack. **When NOT to choose:** Don't choose this if you need a production-ready solution (currently marked as work in progress), or if you're using build tools not yet supported like Vite. ### [easy-mcp](https://mcpcatalogs.com/en/server/zcaceres-easy-mcp) by zcaceres · score 40 · ★194 A TypeScript framework for creating MCP servers with simple Express-like API and experimental decorators. **When to choose:** Choose EasyMCP when you're working with TypeScript and want to quickly create MCP servers with minimal boilerplate code, especially if you prefer a declarative approach with decorators or an Express-like API. **When NOT to choose:** Don't choose EasyMCP if you need production-ready decorator API functionality, require MCP sampling or SSE support, or prefer a more mature and stable library with fewer beta limitations. ### [langchain-mcp-tools-ts](https://mcpcatalogs.com/en/server/hideya-langchain-mcp-tools-ts) by hideya · score 40 · ★23 A TypeScript utility to convert MCP server tools into LangChain-compatible tools with schema adjustments for LLM compatibility. **When to choose:** Choose this when working with TypeScript/LangChain and need to integrate multiple MCP servers with LLM provider-specific schema compatibility. **When NOT to choose:** Consider using LangChain's official MCP adapters if you need comprehensive integration with all MCP features beyond Tools. ### [llms-txt-generator](https://mcpcatalogs.com/en/server/aircodelabs-llms-txt-generator) by aircodelabs · score 40 · ★18 AI-powered generator for llms.txt and llms-full.txt files with MCP server integration. **When to choose:** Choose this when you need AI-optimized documentation for your projects and want to leverage MCP integration for seamless documentation generation within AI development environments. **When NOT to choose:** Avoid if you need extensive customization beyond what the tool offers or if you're not using MCP-compatible AI development environments like Cursor or Claude Desktop. ### [searchcraft-mcp-server](https://mcpcatalogs.com/en/server/searchcraft-inc-searchcraft-mcp-server) by searchcraft-inc · score 40 · ★7 A TypeScript MCP server that integrates Searchcraft's developer search engine for indexing documents, managing search operations, and creating search applications. **When to choose:** Choose this when you need to build search applications with Searchcraft and prefer to manage everything through natural language prompts to Claude. **When NOT to choose:** Avoid this if you're already invested in another search solution, as it creates vendor lock-in to the Searchcraft ecosystem. ### [unity-mcp](https://mcpcatalogs.com/en/server/wondeks-unity-mcp) by wondeks · score 40 · ★21 Unity MCP server enabling AI clients to perform Unity Editor actions through the Model Context Protocol. **When to choose:** Unity developers who want AI assistants to directly manipulate Unity editor objects, assets, and scenes without switching contexts. **When NOT to choose:** If you need extensive customization, complex integration capabilities, or require more detailed documentation of available tools and commands. ### [drawio-architecture-skill](https://mcpcatalogs.com/en/server/moon09300731-drawio-architecture-skill) by moon09300731 · score 40 · ★13 A Claude Code skill for creating architecture diagrams using draw.io via MCP server. **When to choose:** Choose this MCP server when you need to create professional architecture diagrams and technical documentation using Claude Code, especially when working with complex system designs or requiring multiple export formats. **When NOT to choose:** Avoid this if you need real-time collaborative editing capabilities or if you're not already using Claude Code as your primary interface for diagram creation. ### [Claude-Code-MCP-Manager](https://mcpcatalogs.com/en/server/qdhenry-claude-code-mcp-manager) by qdhenry · score 40 · ★24 A bash script tool for managing MCP configurations for Claude Code with add, remove, and auto-loading capabilities. **When to choose:** Choose this tool when you need to manage multiple MCP servers for Claude Code and want an easy way to switch between configurations without manual setup each time. **When NOT to choose:** Don't choose this if you're using a different AI assistant that doesn't support Claude Code's MCP implementation, or if you prefer a GUI-based configuration tool. ### [hasmcp-ce](https://mcpcatalogs.com/en/server/hasmcp-hasmcp-ce) by hasmcp · score 40 · ★27 HasMCP-CE converts API endpoints to MCP servers without code using OpenAPI specs. **When to choose:** Choose HasMCP-CE when you need to quickly expose existing APIs as MCP servers without writing custom MCP implementations, especially if your APIs already have OpenAPI documentation. **When NOT to choose:** Don't choose HasMCP-CE if you need fine-grained control over MCP tool behavior, response optimization, or if you require advanced authentication methods not supported by the community edition. ### [mcp-server-microsoft-paint](https://mcpcatalogs.com/en/server/ghuntley-mcp-server-microsoft-paint) by ghuntley · score 40 · ★20 An MCP server for controlling Microsoft Paint with drawing, shape tools, and window management via JSON-RPC. **When to choose:** When you need programmatic control over Microsoft Paint for simple drawing tasks or educational purposes. **When NOT to choose:** For serious image editing work; this server provides basic functionality only and is limited to Windows with Microsoft Paint installed. ### [heuristic-mcp](https://mcpcatalogs.com/en/server/softerist-heuristic-mcp) by softerist · score 40 · ★21 An MCP server that enables semantic code search with call-graph proximity and recency ranking for AI coding assistants. **When to choose:** Choose this if you're working with JavaScript codebases and need advanced semantic search capabilities that understand code relationships and context. **When NOT to choose:** Not suitable if you need cross-language support, have non-JavaScript codebases, or require offline functionality as it needs internet access for model downloads. ### [mcp-server-any-openapi](https://mcpcatalogs.com/en/server/baryhuang-mcp-server-any-openapi) by baryhuang · score 40 · ★82 An MCP server enabling Claude to discover and call any API endpoint through semantic search of OpenAPI specifications. **When to choose:** Choose this server when you need to integrate private APIs with Claude Desktop and your OpenAPI specifications are too large for standard MCP implementations to handle. **When NOT to choose:** Avoid this if you need to support architectures other than linux/amd64 and linux/arm64, or if you can't tolerate the ~15s cold start penalty for model loading. ### [automcp](https://mcpcatalogs.com/en/server/lirantal-automcp) by lirantal · score 40 · ★5 AutoMCP automatically detects your package.json dependencies and configures MCP servers for your coding agents. **When to choose:** Choose AutoMCP when you have multiple Node.js projects with many dependencies and want to automatically configure relevant MCP servers without manual setup. **When NOT to choose:** Avoid AutoMCP if you need fine-grained control over server configuration, work with non-GitHub dependencies, or prefer manually managing your MCP servers. ### [mcp-delphi](https://mcpcatalogs.com/en/server/flydev-fr-mcp-delphi) by flydev-fr · score 40 · ★17 MCP server for building and cleaning Delphi, Lazarus, and Free Pascal projects via MSBuild and lazbuild tools. **When to choose:** Choose this MCP server when working with Delphi, Lazarus, or Free Pascal projects and need AI assistance with build automation via MCP. **When NOT to choose:** Don't choose this if you're not using Pascal development tools or if you need cross-platform build support beyond Windows. ### [Agentic-MCP-Skill](https://mcpcatalogs.com/en/server/cablate-agentic-mcp-skill) by cablate · score 40 · ★32 A progressive MCP client with three-layer lazy loading that reduces token usage by up to 86%. **When to choose:** Choose this MCP client when working with MCP servers that have many tools but you only need to use occasionally, or when token usage efficiency is a critical concern. **When NOT to choose:** Don't choose this for production environments as it's explicitly labeled as an early concept validation with known issues. Consider mature MCP implementations instead. ### [go-playground-mcp](https://mcpcatalogs.com/en/server/samber-go-playground-mcp) by samber · score 40 · ★5 MCP server for executing Go code in Go Playground and generating shareable URLs. **When to choose:** Choose this MCP server when you need to execute Go code in a sandbox environment or generate shareable links for Go snippets within your AI workflow. **When NOT to choose:** Avoid this if you need to execute code for other programming languages or require persistent code storage beyond the temporary playground environment. ### [mcp-servers](https://mcpcatalogs.com/en/server/dunialabs-mcp-servers) by dunialabs · score 40 · ★6 A collection of 15+ MCP servers for popular services like Google Workspace, GitHub, Notion, and more. **When to choose:** Choose this collection when you need MCP servers for multiple popular APIs in one place, especially when working with Peta Core and wanting integrated security features. **When NOT to choose:** Avoid if you need a highly customized server implementation or if you require advanced features beyond what's provided by the underlying API integrations. ### [roo-logger](https://mcpcatalogs.com/en/server/annenpolka-roo-logger) by annenpolka · score 40 · ★24 An MCP server that automatically logs AI coding assistant activities like command executions and code generation. **When to choose:** Choose this when you need to track and analyze AI coding assistant activities for productivity analysis, context restoration, or documentation purposes. **When NOT to choose:** Avoid if you need real-time monitoring or advanced analytics capabilities, as this server only provides basic JSON logging and search functionality. ### [CodexClaw](https://mcpcatalogs.com/en/server/mackding-codexclaw) by MackDing · score 40 · ★24 Telegram bot for Claude Code and Codex CLI with MCP routing, multi-agent orchestration, cron jobs, and access controls. **When to choose:** Choose this if you need Telegram-based remote access to Codex with clean separation between coding tasks and explicit MCP tool calls, especially when working with multiple Git repositories. **When NOT to choose:** Not suitable if you need a general-purpose MCP server beyond Codex integration, or if you require write access to repositories (this bot only has read capabilities except through explicit GitHub actions). ### [cursor-rust-tools](https://mcpcatalogs.com/en/server/terhechte-cursor-rust-tools) by terhechte · score 40 · ★83 MCP server enabling Rust LLMs to access Rust Analyzer, Crate Docs, and Cargo commands in Cursor. **When to choose:** Developers using Cursor for Rust development who need AI access to real-time type information, documentation, and Cargo commands. **When NOT to choose:** Users not working with Cursor or those needing more comprehensive LSP functionality than what's currently exposed. ### [terminal_server](https://mcpcatalogs.com/en/server/theailanguage-terminal-server) by theailanguage · score 40 · ★83 MCP server that enables AI models to execute terminal commands with multiple setup options. **When to choose:** Choose this MCP server when you need to execute terminal commands through AI assistants in local or cloud environments, particularly when Docker deployment options are preferred. **When NOT to choose:** Avoid this server if you need enhanced security features to block dangerous commands, as the README explicitly notes this as a recommended improvement that hasn't been implemented. ### [create-mcp-app](https://mcpcatalogs.com/en/server/boguan-create-mcp-app) by boguan · score 40 · ★57 CLI tool for scaffolding MCP server applications with TypeScript support and modern development tooling. **When to choose:** Choose this when you need to quickly scaffold new MCP server or client applications with TypeScript and modern tooling. **When NOT to choose:** Don't choose this if you prefer a different language like Python or if you need more specialized MCP functionality not provided by the templates. ### [uMCP](https://mcpcatalogs.com/en/server/mika-f-umcp) by mika-f · score 40 · ★8 uMCP is a lightweight MCP server implementation for Unity applications that uses Streamable HTTP protocol to connect AI agents safely. **When to choose:** Choose uMCP when developing AI-integrated Unity applications and you need a lightweight, dependency-free MCP server specifically designed for game development workflows. **When NOT to choose:** Avoid uMCP if you need MCP server support outside of Unity environments, require authentication mechanisms beyond basic HTTP connections, or need extensive community and third-party tool integrations. ### [bruno-mcp](https://mcpcatalogs.com/en/server/macarthy-bruno-mcp) by macarthy · score 40 · ★23 MCP server for programmatically generating Bruno API testing files, collections, environments, and requests. **When to choose:** When you're already using Bruno for API testing and want to automate the creation of test collections through MCP-compatible AI assistants. **When NOT to choose:** If you're not using Bruno for API testing, or if you need testing capabilities beyond what Bruno offers like performance testing or detailed reporting. ### [mcp-graphql-forge](https://mcpcatalogs.com/en/server/toolprint-mcp-graphql-forge) by toolprint · score 40 · ★18 MCP server that converts GraphQL APIs into MCP tools with automatic parameter validation and field selection caching. **When to choose:** Choose this server when working with GraphQL APIs and need AI tool integration without manual query building. Ideal for teams already using GraphQL who want to expose their APIs to AI assistants. **When NOT to choose:** Don't choose this if you primarily work with REST APIs or need complex query capabilities that go beyond simple introspection-based tool generation. ### [mcpdog](https://mcpcatalogs.com/en/server/kinhunt-mcpdog) by kinhunt · score 40 · ★18 Universal MCP Server Manager providing single interface to configure and manage multiple MCP servers with web dashboard. **When to choose:** Choose MCPDog when you need to manage multiple MCP servers and want a unified interface, web dashboard, and centralized configuration. **When NOT to choose:** Avoid MCPDog if you only need a single MCP server or if you're working in an environment with strict network restrictions that block the dashboard port. ### [penpot-mcp-server](https://mcpcatalogs.com/en/server/zcube-penpot-mcp-server) by zcube · score 39 · ★16 Penpot MCP server connects AI assistants to Penpot design platform with comprehensive design manipulation capabilities. **When to choose:** Choose this MCP server when you need AI assistants to create or manipulate designs in Penpot, especially if you're already using Penpot as your design platform. **When NOT to choose:** Avoid if you need design tool interoperability beyond Penpot, or if you need commercial Penpot cloud features that require authentication beyond access tokens. ### [go-mcp-example](https://mcpcatalogs.com/en/server/rameshsunkara-go-mcp-example) by rameshsunkara · score 39 · ★4 A well-structured Go MCP server implementing analytics reporting with modern Go architecture and enterprise-grade features. **When to choose:** Choose this if you're developing Go-based MCP servers and need a production reference with analytics capabilities. **When NOT to choose:** Don't choose if you need non-analytics MCP functionality or are not using Go as your primary development language. ### [short-url](https://mcpcatalogs.com/en/server/fengzhongsen-short-url) by fengzhongsen · score 39 · ★19 A simple open-source URL shortening service with MCP server for AI assistant integration. **When to choose:** Choose this MCP server when you need a self-hosted, privacy-focused URL shortening solution with direct AI integration capabilities. **When NOT to choose:** Don't choose this if you need a scalable enterprise solution with advanced analytics or if you prefer a cloud-based service without self-hosting requirements. ### [openapi-mcp-swagger](https://mcpcatalogs.com/en/server/salacoste-openapi-mcp-swagger) by salacoste · score 39 · ★16 Converts Swagger/OpenAPI specs into searchable MCP servers for AI assistants, enabling intelligent API documentation queries and code generation. **When to choose:** Choose this when working with complex APIs that exceed AI context windows and need intelligent endpoint discovery with generated code examples. **When NOT to choose:** Avoid if you need real-time API interaction capabilities (this is read-only) or require authentication beyond basic API key support. ### [nitro-mcp-server](https://mcpcatalogs.com/en/server/peterbud-nitro-mcp-server) by peterbud · score 39 · ★5 A lightweight MCP server built with Nitro, featuring HTTP transport, authentication, and example tools/resources. **When to choose:** Choose this when you need a lightweight MCP server with authentication support and prefer a TypeScript implementation using Nitro. **When NOT to choose:** Avoid if you need a server with extensive built-in tools or if you prefer a different transport protocol than HTTP. ### [MCPStack](https://mcpcatalogs.com/en/server/mcp-pipeline-mcpstack) by MCP-Pipeline · score 39 · ★16 MCPStack is a scikit-learn-like pipeline orchestrator for MCP tools, enabling secure composition and chaining of MCP tools in LLM environments. **When to choose:** Choose MCPStack when you need to orchestrate multiple MCP tools into secure, reproducible pipelines for LLM interactions, especially if you're familiar with scikit-learn's API style. **When NOT to choose:** Don't choose MCPStack if you need a single MCP tool, prefer a more traditional MCP server approach, or if you're looking for extensive community support as the project is still relatively new. ### [arcade-mcp-ts](https://mcpcatalogs.com/en/server/arcadeai-arcade-mcp-ts) by ArcadeAI · score 38 · ★2 TypeScript framework for building MCP servers with OAuth, auth, secrets, middleware, and multi-user support. **When to choose:** Choose this framework when building MCP servers with authentication needs, especially if you're already using TypeScript and want a structured way to manage secrets and OAuth integrations. **When NOT to choose:** Avoid if you need a lightweight solution with minimal dependencies or if you're building a simple server without authentication requirements. ### [ace-mcp-go](https://mcpcatalogs.com/en/server/xiaoxu123195-ace-mcp-go) by xiaoxu123195 · score 38 · ★18 Go-based MCP server for codebase indexing and semantic search with Augment compatibility. **When to choose:** Choose Ace-MCP-Go if you need high-performance codebase indexing and semantic search for AI assistants, particularly if you prefer Go-based solutions or need cross-platform support. **When NOT to choose:** Avoid this server if you require write access to codebases (as it's read-only) or if you need advanced search features without relying on an external API for semantic capabilities. ### [offeryn](https://mcpcatalogs.com/en/server/avahowell-offeryn) by avahowell · score 38 · ★37 Rust-based MCP server implementation with tool generation macros and multiple transport options. **When to choose:** Choose offeryn when you need to build MCP servers in Rust, especially when you value the performance and safety guarantees Rust provides. **When NOT to choose:** Don't choose offeryn if you need full MCP feature coverage (resources, prompts, WebSocket) or prefer developing in other languages. ### [code-mcp](https://mcpcatalogs.com/en/server/54yyyu-code-mcp) by 54yyyu · score 38 · ★37 Code-MCP connects Claude AI to dev environments via MCP, enabling terminal commands and file operations through the AI interface. **When to choose:** Choose Code-MCP when you need Claude AI to interact directly with your local or remote development environment through terminal commands and file operations. **When NOT to choose:** Avoid if you need multi-user collaboration features or require more complex workflow automation beyond file operations and command execution. ### [atlas-docs-mcp](https://mcpcatalogs.com/en/server/cartographai-atlas-docs-mcp) by CartographAI · score 38 · ★35 An MCP server providing clean markdown documentation for popular libraries and frameworks to enhance LLM code generation accuracy. **When to choose:** Choose this MCP server when working with less popular or newly released libraries that LLMs may not have been trained on, and you need reliable, up-to-date documentation references. **When NOT to choose:** Don't choose this if you need documentation for very niche libraries not in their supported list, or if you require offline access to documentation. ### [mcpc](https://mcpcatalogs.com/en/server/micl2e2-mcpc) by micl2e2 · score 38 · ★27 A cross-platform C SDK for MCP implementing tools, resources, prompts and completion with modern C23. **When to choose:** Choose mcpc when building performance-critical MCP servers or clients where native C execution and minimal dependencies are priorities, especially for embedded systems or cross-platform deployment without runtime dependencies. **When NOT to choose:** Avoid mcpc if you're developing rapid prototypes or need higher-level abstractions, as it requires manual memory management and doesn't provide the convenience of Python or JavaScript SDKs. ### [mcp-java-sdk-examples](https://mcpcatalogs.com/en/server/thought2code-mcp-java-sdk-examples) by thought2code · score 38 · ★32 A collection of MCP server examples demonstrating different Java SDK implementations. **When to choose:** Choose this if you're working with Java applications and need practical examples of implementing MCP servers with different SDK approaches. **When NOT to choose:** Not ideal if you need documentation beyond examples or if you're working with non-Java environments. ### [BitbucketMcpServers](https://mcpcatalogs.com/en/server/peakflames-bitbucketmcpservers) by peakflames · score 38 · ★3 C# implementation of MCP server for Bitbucket integration with pull request operations. **When to choose:** Choose this if you're already using Bitbucket and want to integrate pull request management into AI assistants like Cline, especially when you prefer .NET/C# stack or need both remote server and local console deployment options. **When NOT to choose:** Avoid if you're using GitHub instead of Bitbucket, or if you need more extensive repository management features beyond pull requests (commits, branches, etc.). ### [basecoat-ui-mcp](https://mcpcatalogs.com/en/server/sorbh-basecoat-ui-mcp) by Sorbh · score 38 · ★9 MCP server providing Basecoat CSS components and documentation for AI-assisted HTML development. **When to choose:** Choose this when you're working with Basecoat CSS or ShadCN UI and need AI assistance to discover components, get documentation, or generate HTML interfaces. **When NOT to choose:** Don't choose this if you're not using Basecoat CSS, as the server provides no value for other CSS frameworks. Consider general-purpose UI component servers instead. ### [mcp-local-dev](https://mcpcatalogs.com/en/server/txbm-mcp-local-dev) by txbm · score 38 · ★28 AI-powered local development environment manager that handles setup, dependency management, and testing through MCP. **When to choose:** Choose this when you want AI to handle environment setup and testing for local development, especially when working with multiple programming languages or unfamiliar codebases. **When NOT to choose:** Avoid if you need extensive customization of development environments or are working in highly restricted environments without network access. ### [MCPify](https://mcpcatalogs.com/en/server/abdebek-mcpify) by abdebek · score 38 · ★4 MCPify is a .NET library that converts OpenAPI/Swagger specs into MCP tools for AI assistants. **When to choose:** Choose MCPify if you're already using .NET/ASP.NET Core and want to expose existing APIs or OpenAPI specifications as MCP tools with robust OAuth support. **When NOT to choose:** Don't choose MCPify if you need to expose non-REST APIs or if you're not working in the .NET ecosystem. ### [mcp_autogen_sse_stdio](https://mcpcatalogs.com/en/server/sam-92-mcp-autogen-sse-stdio) by SaM-92 · score 38 · ★29 A practical example demonstrating AutoGen integration with both local and remote MCP servers using Stdio and SSE. **When to choose:** Choose this when you need to integrate AutoGen with both local and remote MCP tools, particularly when working with hybrid architectures combining simple calculation tools with external web services. **When NOT to choose:** Not suitable for production environments requiring extensive tool integration as this is a demonstration project focused on showing protocol capabilities rather than providing a comprehensive solution. ### [mcp-congress_gov_server](https://mcpcatalogs.com/en/server/bsmi021-mcp-congress-gov-server) by bsmi021 · score 38 · ★8 MCP server providing access to U.S. Congress.gov legislative data through hybrid resources and tools. **When to choose:** Choose this server when you need to integrate U.S. legislative data into AI applications that can benefit from structured access to Congress.gov information with proper rate limiting and hybrid resource/tool approach. **When NOT to choose:** Don't choose this if you need non-U.S. legislative data, as it only provides access to U.S. Congress information through the Congress.gov API. ### [dockashell](https://mcpcatalogs.com/en/server/anzax-dockashell) by anzax · score 38 · ★30 DockaShell provides AI agents with isolated Docker containers for persistent shell access, file operations, and audit trails. **When to choose:** Choose DockaShell when you need AI agents to have persistent environments and full shell capabilities for complex, multi-step tasks that require maintaining state across sessions. **When NOT to choose:** Avoid DockaShell for production systems due to its research status and expect changes. It's not ideal for high-security environments or when you need strict command validation. ### [mcp-manager](https://mcpcatalogs.com/en/server/nstebbins-mcp-manager) by nstebbins · score 38 · ★25 MCP Manager CLI tool simplifies installation and management of MCP servers across Claude Desktop, Cursor and other clients. **When to choose:** Choose this tool if you're using multiple MCP clients and need a unified way to discover, install, and manage servers across them. **When NOT to choose:** Don't choose this if you only use a single MCP client or prefer manual configuration of servers. ### [mcp-template-node](https://mcpcatalogs.com/en/server/rethunk-ai-mcp-template-node) by Rethunk-AI · score 37 · ★1 TypeScript template for Node.js MCP servers with a notes management example. **When to choose:** Developers looking to create MCP servers in Node.js/TypeScript who want a comprehensive template with complete examples and testing infrastructure. **When NOT to choose:** You need production-ready persistent storage (this template uses in-memory storage) or are looking for a template in a different language. ### [keyshade-mcp](https://mcpcatalogs.com/en/server/keyshade-xyz-keyshade-mcp) by keyshade-xyz · score 37 · ★8 Keyshade MCP server provides secure API access to manage secrets and keys within development environments. **When to choose:** Choose this if you're already using Keyshade services and want to integrate them with Claude Desktop or VSCode through MCP. **When NOT to choose:** Don't choose this if you're not using Keyshade services or need a more comprehensive API integration solution. ### [mobile-mcp](https://mcpcatalogs.com/en/server/runablehq-mobile-mcp) by runablehq · score 37 · ★66 An MCP server that enables LLMs to automate Android mobile devices through structured UI interaction. **When to choose:** Choose Mobile MCP when you need to automate interactions with Android apps or devices without visual dependencies, particularly for testing or UI automation tasks. **When NOT to choose:** Don't choose Mobile MCP if you need iOS support, visual-based automation, or require more comprehensive mobile device management capabilities beyond UI interaction. ### [grumpydev-mcp](https://mcpcatalogs.com/en/server/sinedied-grumpydev-mcp) by sinedied · score 37 · ★21 An MCP server that simulates a grumpy senior developer reviewing code with sarcastic feedback. **When to choose:** Choose this for team environments where a lighthearted, entertaining approach to code review is desired to maintain developer morale while still getting valuable feedback. **When NOT to choose:** Avoid in professional environments requiring formal, diplomatic communication or when the sarcastic tone might be perceived as unprofessional or demotivating to team members. ### [mcp-config](https://mcpcatalogs.com/en/server/marcusschiesser-mcp-config) by marcusschiesser · score 37 · ★64 A CLI tool for easy installation and management of MCP server configurations. **When to choose:** Choose MCP Config when you need a simple, command-line way to manage multiple MCP server installations and configurations. **When NOT to choose:** Don't choose this if you need GUI-based configuration management or support for MCP clients beyond Claude, Windsurf, and Cursor. ### [aisdk-mcp-bridge](https://mcpcatalogs.com/en/server/vrknetha-aisdk-mcp-bridge) by vrknetha · score 37 · ★23 Bridge package connecting MCP servers with AI SDK tools, supporting multiple server types and real-time communication. **When to choose:** Choose this bridge if you're integrating AI models with multiple MCP servers and want a clean, TypeScript-native solution with real-time communication support. **When NOT to choose:** Avoid if you need only a single MCP server integration or require authentication beyond environment variables, as the bridge doesn't handle complex auth models. ### [mcp-server-codegraph](https://mcpcatalogs.com/en/server/cartographai-mcp-server-codegraph) by CartographAI · score 37 · ★20 MCP server that transforms codebases into interactive knowledge graphs with entity tracking and relationship mapping. **When to choose:** Choose this server when you need to understand complex codebase relationships across multiple languages and want to leverage AI for code analysis and refactoring. **When NOT to choose:** Avoid this if you need real-time code analysis or are working with very large codebases that might exceed the graph's performance limits. ### [mcp](https://mcpcatalogs.com/en/server/emillindfors-mcp) by EmilLindfors · score 37 · ★18 Rust implementation of MCP providing standardized communication between AI models and external resources. **When to choose:** Choose mcp.rs when developing Rust-based AI applications requiring high performance and type safety for resource access. **When NOT to choose:** Avoid if you need language support beyond Rust or require extensive documentation, as the project appears to be in early development with limited examples. ### [zero-mcp](https://mcpcatalogs.com/en/server/sniko-zero-mcp) by SNIKO · score 37 · ★2 Zero MCP is a lightweight, zero-boilerplate toolkit for building fast HTTP-based MCP servers with minimal dependencies. **When to choose:** Choose Zero MCP when you need a fast, lightweight HTTP-based MCP server with minimal dependencies and want to avoid the Express-based overhead of the official SDK. **When NOT to choose:** Don't choose Zero MCP if you need full MCP specification support including stdio transports, prompts API, or authentication flows, as it only supports HTTP transport. ### [shadcnspace-mcp](https://mcpcatalogs.com/en/server/shadcnspace-shadcnspace-mcp) by shadcnspace · score 37 · ★1 MCP server for discovering and installing ShadcnSpace UI blocks directly in AI-powered IDEs. **When to choose:** Choose this when you're working on UI projects with ShadcnSpace components and want AI assistance to discover and integrate blocks directly in your IDE. **When NOT to choose:** Not ideal if you're not using ShadcnSpace components or if your IDE doesn't support MCP integration. ### [jiraxmcp](https://mcpcatalogs.com/en/server/vaibhavpandeyvpz-jiraxmcp) by vaibhavpandeyvpz · score 37 · ★1 Jira MCP server providing issue management, agile tools, and webhook notifications through the Model Context Protocol. **When to choose:** Choose jiraxmcp when you need MCP-compatible AI access to Jira Cloud, especially if you want to create issues, manage agile workflows, or receive webhook notifications through MCP. **When NOT to choose:** Don't choose this if you need Jira server (non-Cloud) integration, as it's specifically designed for Jira Cloud instances only. ### [mcpifier](https://mcpcatalogs.com/en/server/summerdawn-ai-mcpifier) by summerdawn-ai · score 37 · ★1 Mcpifier is a zero-code gateway that converts REST APIs into MCP servers using Swagger/OpenAPI specs. **When to choose:** Choose Mcpifier when you need to quickly expose existing REST APIs as MCP tools without modifying the API itself, especially if you're already working in a .NET ecosystem. **When NOT to choose:** Avoid Mcpifier if you need streaming support, binary data handling, or authentication beyond what your existing REST API provides, as it currently lacks these features. ### [mcp_mediator](https://mcpcatalogs.com/en/server/makbn-mcp-mediator) by makbn · score 37 · ★10 Java-based framework that automatically generates MCP servers from existing source code, services, and external tools. **When to choose:** Choose this if you're working with Java codebases and want to quickly convert existing services to MCP servers without modifying your code, or need to aggregate multiple MCP servers behind a single interface. **When NOT to choose:** Avoid this if you need production-ready features, comprehensive documentation, or are not working with Java/Spring ecosystems, as many features are still in progress. ### [redm-mcp-public](https://mcpcatalogs.com/en/server/cmoen11-redm-mcp-public) by Cmoen11 · score 37 · ★1 MCP server providing RedM/RDR3 native lookups, semantic search and framework documentation. **When to choose:** Choose this MCP server if you're developing for RedM/RDR3 and need precise native function lookups and framework documentation within your AI coding environment. **When NOT to choose:** Avoid this if you need access to the server source code (it's proprietary) or if you're working with game development platforms other than RedM/RDR3. ### [mcp-server-python-template](https://mcpcatalogs.com/en/server/sontallive-mcp-server-python-template) by sontallive · score 37 · ★17 Python template for building MCP servers with clear documentation and example implementations. **When to choose:** Choose this template when you're starting with MCP server development in Python and want a well-documented foundation with example implementations. **When NOT to choose:** Don't choose this if you need production-ready features beyond what's provided in the template, or if you prefer other programming languages. ### [OCTALUME](https://mcpcatalogs.com/en/server/harery-octalume) by Harery · score 37 · ★1 AI-native SDLC framework with 8-phase gates, compliance scanners, and MCP integration for regulated industries. **When to choose:** Choose OCTALUME if you're in regulated industries (healthcare, finance, government) requiring auditable AI-driven development with built-in compliance tracking. **When NOT to choose:** Avoid OCTALUME for small projects without compliance requirements, or if you prefer traditional development approaches without AI agents. ### [mcp-go](https://mcpcatalogs.com/en/server/xiaoconstantine-mcp-go) by XiaoConstantine · score 37 · ★10 Go implementation of MCP protocol with client and server components, supporting stdio and HTTP transports. **When to choose:** Choose MCP-Go if you're developing Go applications that need to implement MCP servers or clients, especially when you require control over transport protocols or need to integrate with existing Go-based infrastructure. **When NOT to choose:** Don't choose MCP-Go if you're working primarily in other languages, as there are more mature implementations available for Python and JavaScript ecosystems. ### [mcp-csharp-starter](https://mcpcatalogs.com/en/server/sammorrowdrums-mcp-csharp-starter) by SamMorrowDrums · score 36 · ★0 A feature-complete MCP server template in C# with tools, resources, and prompts using .NET 8. **When to choose:** Choose this if you're developing .NET applications and want to integrate MCP functionality using idiomatic C# code. **When NOT to choose:** Avoid if you're not using .NET/C# or need a solution that's not tied to Microsoft's ecosystem. ### [go-mcp](https://mcpcatalogs.com/en/server/dstotijn-go-mcp) by dstotijn · score 36 · ★16 A Go library for implementing MCP servers with protocol support, type-safe handlers, and validation. **When to choose:** Choose go-mcp if you're developing Go applications that need to implement MCP servers or clients with type safety and built-in validation. **When NOT to choose:** Avoid if you need extensive documentation or examples beyond the basic server example provided in the repository. ### [go-mcp](https://mcpcatalogs.com/en/server/megagrindstone-go-mcp) by MegaGrindStone · score 36 · ★46 Go implementation of MCP protocol with server and client functionality, supporting tools, resources, and prompts. **When to choose:** Choose go-mcp when you need to build MCP servers or clients in Go, especially for systems requiring pluggable transport options like SSE or Standard IO. **When NOT to choose:** Avoid go-mcp if you need stability, as the main branch is noted to contain unreleased changes that may be unstable. Consider alternatives if you require a more mature implementation. ### [aica](https://mcpcatalogs.com/en/server/dotneet-aica) by dotneet · score 36 · ★16 aica is an open-source AI code analyzer that supports MCP server functionality for code review and analysis. **When to choose:** Choose aica if you need an open-source, customizable AI code analysis tool that can be integrated as an MCP server and works with your preferred LLM provider. **When NOT to choose:** Avoid aica if you need a commercial tool with official support or require advanced IDE integration beyond what MCP provides. ### [user-feedback-mcp](https://mcpcatalogs.com/en/server/mrexodia-user-feedback-mcp) by mrexodia · score 36 · ★53 A simple MCP server enabling human-in-the-loop workflows for AI development tools like Cline and Cursor. **When to choose:** Choose this when developing desktop applications that need user feedback during development, especially when using Cline or Cursor as your AI assistant. **When NOT to choose:** Not suitable for web applications or APIs, and lacks authentication - don't expose to untrusted environments. ### [oatpp-mcp](https://mcpcatalogs.com/en/server/oatpp-oatpp-mcp) by oatpp · score 36 · ★49 C++ MCP server implementation for Oat++ framework with automated API tool generation. **When to choose:** Choose oatpp-mcp if you're already using Oat++ for C++ development and want to expose your API endpoints as MCP tools to LLMs. **When NOT to choose:** Not suitable for non-C++ projects or those requiring transports other than STDIO/SSE; requires knowledge of Oat++ framework. ### [mcp-server-pipecd](https://mcpcatalogs.com/en/server/warashi-mcp-server-pipecd) by Warashi · score 36 · ★4 MCP server for PipeCD, a continuous deployment platform, providing application management through the Model Context Protocol. **When to choose:** Teams already using PipeCD who want to integrate deployment management into their AI workflow or chat interfaces. **When NOT to choose:** Not suitable for organizations not using PipeCD, as this server provides no value without the underlying platform. ### [connector](https://mcpcatalogs.com/en/server/opticlm-connector) by OpticLM · score 36 · ★3 Abstract interface for LLMs to connect to code analysis tools via MCP and AI SDK implementations. **When to choose:** Choose this connector when you need to integrate LLM capabilities with IDE environments that support LSP, particularly for code navigation, diagnostics, and editing features. **When NOT to choose:** Avoid this if you need a general-purpose MCP server for non-code-related tasks or if you're looking for a solution with broader language support beyond what LSP provides. ### [mcp-inspector-desktop](https://mcpcatalogs.com/en/server/cicbyte-mcp-inspector-desktop) by cicbyte · score 36 · ★0 A native desktop wrapper for MCP Inspector built with Tauri v2 that provides a seamless debugging experience without browser tabs. **When to choose:** Choose this when you want a dedicated desktop interface for debugging MCP servers without switching between browser tabs and your development environment. **When NOT to choose:** Don't choose this if you prefer web-based debugging tools or if you need advanced features not provided by MCP Inspector itself. ### [k6-mcp-server](https://mcpcatalogs.com/en/server/qainsights-k6-mcp-server) by QAInsights · score 36 · ★24 k6 MCP server for load testing integration with AI models through Model Context Protocol. **When to choose:** Choose this server if you're already using k6 for performance testing and want to integrate it into your AI-powered development workflow for automated test execution and analysis. **When NOT to choose:** Don't choose this if you need advanced k6 scripting features not exposed by the MCP server, or if you're already using a different load testing tool that better fits your workflow. ### [win-cli-mcp-server](https://mcpcatalogs.com/en/server/mhprol-win-cli-mcp-server) by mhprol · score 36 · ★0 A hardened MCP server for Windows CLI and SSH interactions with critical bug fixes and security improvements. **When to choose:** Choose this fork when you need reliable Windows CLI access through MCP and the original project has issues with stability or security. **When NOT to choose:** Avoid if you're working on non-Windows systems or need cross-platform CLI access, as this is specifically designed for Windows environments. ### [create-mcp](https://mcpcatalogs.com/en/server/fefergrgrgrg-create-mcp) by fefergrgrgrg · score 36 · ★16 CLI tool to create and deploy MCP servers to Cloudflare Workers using TypeScript functions with JSDoc comments. **When to choose:** Choose this when you want to quickly prototype or deploy MCP tools without managing local infrastructure, especially if you're already using Cloudflare services. **When NOT to choose:** Avoid this if you need fine-grained control over your MCP server environment or require features not available in Cloudflare Workers. ### [mcp-icon](https://mcpcatalogs.com/en/server/hustcc-mcp-icon) by hustcc · score 36 · ★0 An MCP server providing semantic search for over 100,000 SVG icons via keyword queries. **When to choose:** Choose mcp-icon when you need to quickly find semantic icon matches for UI design, content creation, or development projects without leaving your AI environment. **When NOT to choose:** Don't choose mcp-icon if you need icons with specific licenses, require extensive filtering options beyond keyword search, or need to host your own icon library. ### [django-mcp](https://mcpcatalogs.com/en/server/hyperb1iss-django-mcp) by hyperb1iss · score 36 · ★16 A Python package that bridges Django apps with AI assistants using MCP through decorators. **When to choose:** Choose Django-MCP if you're building a Django application and want to enable AI assistants to interact with your models, admin functions, and custom tools without complex integrations. **When NOT to choose:** Avoid Django-MCP if you need write access to your database (currently read-only), have non-Django applications, or require a production-ready solution (this is still in alpha status). ### [mcp-partner.github.io](https://mcpcatalogs.com/en/server/mcp-partner-mcp-partner-github-io) by mcp-partner · score 36 · ★1 A Postman-like tool for testing and debugging MCP servers via SSE or HTTP. **When to choose:** Choose MCP Partner when you need a quick, no-install web tool to test MCP servers during development or demonstrations, especially when you need to visualize tool interactions and request/response flows. **When NOT to choose:** Don't choose MCP Partner if you need a persistent client with authentication management, comprehensive request history across sessions, or if you're working with strictly internal networks that block external proxy services. ### [otel-server](https://mcpcatalogs.com/en/server/cruz-andr-otel-server) by cruz-andr · score 36 · ★0 MCP server for Docker container inspection and OpenTelemetry trace querying via Jaeger with built-in security features. **When to choose:** Choose this server if you're managing Docker containers and distributed systems with OpenTelemetry and need an AI assistant to help with observability and troubleshooting. **When NOT to choose:** Don't choose this if you don't use Docker or OpenTelemetry, as its functionality is limited to these ecosystems. ### [mcp-starter-template](https://mcpcatalogs.com/en/server/git-fg-mcp-starter-template) by Git-Fg · score 36 · ★0 A TypeScript template for creating autonomous MCP servers with tools-first architecture. **When to choose:** Choose this template when you're building MCP servers specifically for autonomous AI agents with a focus on well-documented tools and their orchestration. **When NOT to choose:** Avoid if you need extensive resources or prompts capabilities, as this template emphasizes tools-first architecture and discourages complex primitives. ### [mcp-explorer](https://mcpcatalogs.com/en/server/ventz-mcp-explorer) by ventz · score 36 · ★1 A web-based GUI for exploring MCP servers, providing an interface to browse tools, resources, and prompts. **When to choose:** When you need a visual way to explore MCP server capabilities without writing code, especially for testing or debugging purposes. **When NOT to choose:** If you need programmatic access to MCP servers or require advanced features beyond exploration and basic testing. ### [dynamic-shell-server](https://mcpcatalogs.com/en/server/codelion-dynamic-shell-server) by codelion · score 36 · ★41 A Python MCP server for secure shell command execution with dynamic approval and audit logging. **When to choose:** Choose this when you need to safely execute shell commands through an AI assistant but want granular control over what commands can run and maintain an audit trail. **When NOT to choose:** Not for production environments requiring high-performance command execution or complex permission models, as it's designed for interactive approval in AI assistant contexts. ### [codenav](https://mcpcatalogs.com/en/server/ajacobm-codenav) by ajacobm · score 36 · ★1 CodeNavigator is an MCP server providing comprehensive code analysis across 25+ programming languages. **When to choose:** Choose CodeNavigator when you need comprehensive code analysis across multiple programming languages with advanced navigation, dependency mapping, and code quality assessment features. **When NOT to choose:** Consider alternatives if you're looking for simple file browsing functionality or need to analyze extremely large codebases (>10,000 files) due to potential performance limitations. ### [mcp](https://mcpcatalogs.com/en/server/sockeon-mcp) by sockeon · score 35 · ★2 PHP-based MCP server providing modular tools, resources, and prompts for Sockeon framework development. **When to choose:** Choose this MCP server if you're developing with the Sockeon framework and need integrated tools for scaffolding, security, and real-time features. **When NOT to choose:** Avoid this server if you're not using the Sockeon framework or prefer solutions with broader language support beyond PHP. ### [mcp-node-omnibus-server](https://mcpcatalogs.com/en/server/bsmi021-mcp-node-omnibus-server) by bsmi021 · score 35 · ★9 A comprehensive MCP server providing Node.js development tooling and automation capabilities. **When to choose:** Choose this server when you're working with Node.js projects and want AI assistance for scaffolding, code generation, and development tasks without switching contexts. **When NOT to choose:** Avoid if you primarily work with non-Node.js stacks or require sophisticated documentation generation beyond what Node.js projects typically need. ### [elysia-mcp-starter](https://mcpcatalogs.com/en/server/kerlos-elysia-mcp-starter) by kerlos · score 35 · ★8 A starter template for building MCP servers with Elysia and Bun runtime. **When to choose:** Teams that want to build MCP servers quickly with TypeScript, especially those already familiar with ElysiaJS. **When NOT to choose:** If you need a more mature framework with built-in authentication, scalability features, or extensive tooling beyond what Elysia provides. ### [adonis-mcp](https://mcpcatalogs.com/en/server/7nohe-adonis-mcp) by 7nohe · score 35 · ★10 AdonisJS package for building remote MCP servers with Server-Sent Events support. **When to choose:** Choose this when building MCP servers within existing AdonisJS applications and prefer using Server-Sent Events for real-time communication. **When NOT to choose:** Avoid this if you're not using AdonisJS framework, as this MCP server is tightly coupled to the AdonisJS ecosystem. ### [mcp-connector](https://mcpcatalogs.com/en/server/foobara-mcp-connector) by foobara · score 35 · ★2 A Ruby gem that exposes Foobara commands as MCP tools for integration with AI assistants like Claude. **When to choose:** Choose this server if you're using Foobara in your Ruby application and want to expose your business logic as MCP tools to AI assistants. **When NOT to choose:** Avoid this server if you're not using Foobara, need advanced features like authentication, or require multi-language support beyond Ruby. ### [mcp-starter](https://mcpcatalogs.com/en/server/instructa-mcp-starter) by instructa · score 35 · ★37 A TypeScript starter template for building MCP servers with support for stdio, HTTP, and SSE protocols. **When to choose:** Choose this starter if you're building a new MCP server from scratch and want TypeScript support with multiple protocol options. **When NOT to choose:** Don't choose this if you need advanced features beyond a basic starter template, or if you prefer more mature frameworks with built-in tool implementations. ### [Glee](https://mcpcatalogs.com/en/server/gleemcp-glee) by GleeMCP · score 35 · ★3 Glee is an MCP toolkit providing memory, code review, and context tools for developers to enhance Claude's capabilities. **When to choose:** Choose Glee when you want a comprehensive, all-in-one MCP toolkit for memory management, code review, and session hooks without needing multiple servers. **When NOT to choose:** Avoid Glee if you need highly specialized tools not covered by its current 11 implemented features or require more stable documentation as it's still in active development. ### [drawio-architecture-skill](https://mcpcatalogs.com/en/server/quant-chiggz-drawio-architecture-skill) by quant-chiggz · score 35 · ★2 A Claude Code skill for creating architecture diagrams using draw.io with customizable styles and multi-format export. **When to choose:** Choose this MCP skill if you need to create architecture diagrams within Claude Code and prefer a simple command-driven interface over traditional drawing tools. **When NOT to choose:** Avoid this skill if you require advanced diagram features, collaborative editing, or if you're working in an environment without Claude Code access. ### [easymcp](https://mcpcatalogs.com/en/server/promptmesh-easymcp) by promptmesh · score 35 · ★18 A high-performance Python SDK for MCP clients with server management, caching, and parallel request support. **When to choose:** Choose this when building Python applications that need to interact with MCP servers and want a high-level API with automatic caching and namespace management. **When NOT to choose:** Don't choose this if you're looking to create an MCP server or need a language other than Python. ### [strava-mcp](https://mcpcatalogs.com/en/server/kw510-strava-mcp) by kw510 · score 35 · ★26 MCP server with Strava OAuth integration for secure authentication and tool access. **When to choose:** Choose this when you need secure OAuth integration with Strava for AI tools like Claude or Cursor, especially if you're already using Cloudflare infrastructure. **When NOT to choose:** Not ideal if you need to work with other fitness platforms besides Strava or require extensive customization beyond the OAuth flow. ### [go-modelcontextprotocol](https://mcpcatalogs.com/en/server/warashi-go-modelcontextprotocol) by Warashi · score 35 · ★3 A pure Go implementation of the Model Context Protocol without third-party dependencies, supporting tools, resources, and multiple transport mechanisms. **When to choose:** Choose this implementation when you're building Go-based AI/LLM backends and prefer a dependency-free, pure Go solution with direct control over MCP implementation details. **When NOT to choose:** Avoid this if you need language bindings for Python, JavaScript, or other languages, or if you require extensive ecosystem integrations beyond the core MCP specification. ### [illustrator-mcp-server](https://mcpcatalogs.com/en/server/spencerhhubert-illustrator-mcp-server) by spencerhhubert · score 35 · ★50 Adobe Illustrator MCP server allowing AI models to run JavaScript scripts directly in Illustrator. **When to choose:** Choose this when you need to automate Adobe Illustrator workflows through AI agents and are exclusively working on macOS. **When NOT to choose:** Don't choose this if you need cross-platform support, Windows compatibility, or require integration with design tools other than Illustrator. ### [shaderc-vkrunner-mcp](https://mcpcatalogs.com/en/server/mehmetoguzderin-shaderc-vkrunner-mcp) by mehmetoguzderin · score 34 · ★6 A toy MCP server for AI agents to safely code, compile, and run GPU shaders using Vulkan via CPU emulation. **When to choose:** Choose this when you need to develop and test GPU shaders without physical hardware, particularly for educational purposes or prototyping complex shader algorithms in a safe environment. **When NOT to choose:** Don't choose this if you need to test with actual GPU hardware, require high-performance shader execution, or need production-ready visualization tools - this is primarily an educational/development tool. ### [MCP.DevOps](https://mcpcatalogs.com/en/server/jihadkhawaja-mcp-devops) by jihadkhawaja · score 34 · ★1 Comprehensive MCP DevOps server for cloud deployments, Docker operations, and infrastructure automation across AWS, Azure, and Linux servers. **When to choose:** Choose this MCP server if you're developing .NET applications and need comprehensive deployment automation across AWS, Azure, and Linux servers with extensive DevOps tooling. **When NOT to choose:** Avoid this if you primarily use non-.NET technologies, need deployment support for platforms other than AWS/Azure/Linux, or require extensive community support and third-party integrations. ### [mcp.mbt](https://mcpcatalogs.com/en/server/colmugx-mcp-mbt) by colmugx · score 34 · ★2 Type-safe MCP server implementation in MoonBit with dual transport support (STDIO/HTTP). **When to choose:** Choose this if you're developing in MoonBit and need a type-safe MCP server implementation with support for both STDIO and HTTP transports. **When NOT to choose:** Avoid this if you need production-ready features like structured logging, metrics collection, or authentication, as these are not yet implemented. ### [openapi2mcptools](https://mcpcatalogs.com/en/server/2013xile-openapi2mcptools) by 2013xile · score 34 · ★19 Converts OpenAPI specifications to MCP tools for faster MCP server development. **When to choose:** Choose this tool when you need to quickly convert existing REST API documentation into MCP tools without implementing the protocol details manually. **When NOT to choose:** Avoid this if you need advanced API features like webhooks, complex authentication flows, or real-time streaming endpoints that go beyond standard REST API patterns. ### [mcp-chain-of-draft-server](https://mcpcatalogs.com/en/server/bsmi021-mcp-chain-of-draft-server) by bsmi021 · score 34 · ★24 MCP server implementing Chain of Draft protocol for systematic reasoning and decision-making in development workflows. **When to choose:** Choose when you need structured, iterative reasoning for complex development decisions and architecture planning. **When NOT to choose:** Not ideal for simple tasks where quick decisions are needed; better suited for complex, multi-stage development processes. ### [Jira_mcp_streamlit](https://mcpcatalogs.com/en/server/pawankumar94-jira-mcp-streamlit) by pawankumar94 · score 34 · ★5 Streamlit-based Jira assistant with MCP integration for creating, searching, and managing tickets. **When to choose:** Choose this server if you need a user-friendly interface for Jira ticket management through natural language commands and prefer direct form-based interactions over API calls. **When NOT to choose:** Don't choose this if you need programmatic access to Jira without a UI, or if your workflow requires integration with other systems beyond what MCP provides. ### [istio-mcp-server](https://mcpcatalogs.com/en/server/krutsko-istio-mcp-server) by krutsko · score 34 · ★1 MCP server providing read-only access to Istio service mesh resources in Kubernetes clusters. **When to choose:** Choose this MCP server when you need AI assistants to query and analyze Istio service mesh configurations without risking accidental modifications to production environments. **When NOT to choose:** Avoid if you need write access to Istio configurations, as this server is strictly read-only. Also not suitable for organizations not using Istio in their Kubernetes environments. ### [framesmith](https://mcpcatalogs.com/en/server/robdavenport-framesmith) by RobDavenport · score 34 · ★1 Fighting game character editor with MCP server for LLM-assisted workflows **When to choose:** Choose this MCP server if you're developing fighting games and want LLM assistance in character creation, balancing, or documentation workflows. **When NOT to choose:** Don't choose this for non-fighting game projects, as the tools and data formats are specialized for fighting game mechanics. ### [mcp-workbench](https://mcpcatalogs.com/en/server/conductor-oss-mcp-workbench) by conductor-oss · score 34 · ★2 A visual debugger and test client for MCP servers with real-time inspection and tool execution capabilities. **When to choose:** Choose this when you need to debug MCP servers visually or manually test protocol interactions during development. **When NOT to choose:** Don't use this in production environments - it's a debugging tool with limited CORS handling and not designed for production workloads. ### [UIUX-MCP](https://mcpcatalogs.com/en/server/re-rank-uiux-mcp) by re-rank · score 34 · ★1 KRDS UI/UX MCP Server integrates Korea's government design system with AI assistants, providing component search, design tokens, and code validation. **When to choose:** Choose this MCP server if you're developing Korean government digital services using KRDS, or if you need to validate your code against KRDS design guidelines. **When NOT to choose:** Not suitable if you're not working with KRDS or Korean government projects, as this server is specifically tailored to Korea's design system. ### [MCP-Manager-GUI](https://mcpcatalogs.com/en/server/gabrielbacha-mcp-manager-gui) by gabrielbacha · score 34 · ★17 A GUI tool for managing MCP server configurations across different clients. **When to choose:** Choose MCP Manager when you need a simple graphical interface to manage MCP configurations across multiple clients without manually editing JSON files. **When NOT to choose:** Don't choose this if you prefer command-line tools or need advanced configuration options not available in a GUI interface. ### [koncept-mcp](https://mcpcatalogs.com/en/server/yourtechtribe-labs-koncept-mcp) by yourtechtribe-labs · score 33 · ★0 MCP server for semantic concept graphs in codebases to capture cross-cutting meanings. **When to choose:** Teams needing to document and enforce high-level semantic constraints across codebases that traditional code analysis tools miss. **When NOT to choose:** For production use as this is pre-alpha with unstable APIs; consider alternatives for critical projects. ### [swift-context-protocol](https://mcpcatalogs.com/en/server/1amageek-swift-context-protocol) by 1amageek · score 33 · ★11 Swift-based MCP server implementing distributed actors for type-safe tools, resources, and prompts in AI contexts. **When to choose:** Choose this MCP server if you're working in a Swift ecosystem and need type-safe distributed AI tools with Swift's actor model. **When NOT to choose:** Avoid this if you need broader language support or if your team isn't using Swift 6+, as distributed actors are a recent Swift feature. ### [mcp_review_code_tool](https://mcpcatalogs.com/en/server/wenkil-mcp-review-code-tool) by wenkil · score 33 · ★13 MCP server for code review using OpenAI API to analyze and review code files. **When to choose:** Choose this if you need a simple web interface for code review and basic MCP integration without complex setup. **When NOT to choose:** Don't choose this if you need a robust MCP server with multiple tools or extensive code analysis capabilities. ### [1mcp](https://mcpcatalogs.com/en/server/slzcdhd-1mcp) by slzcdhd · score 33 · ★1 1mcp is a central proxy server that aggregates multiple MCP servers into a unified interface. **When to choose:** Choose 1mcp when you need to manage multiple MCP servers and provide a unified interface to clients, especially when working with diverse transports or requiring dynamic configuration updates. **When NOT to choose:** Avoid 1mcp for simple use cases with only one MCP server, or if you need fine-grained control over individual server configurations as the abstraction might limit customization options. ### [x-mcp](https://mcpcatalogs.com/en/server/cs-tao-x-mcp) by CS-Tao · score 33 · ★1 A TypeScript-based MCP framework with middleware and mod architecture for extending functionality. **When to choose:** Choose x-mcp when building custom MCP servers with shared middleware capabilities and modular architecture, especially if you're developing complex applications with layered processing. **When NOT to choose:** Avoid x-mcp if you need a simple MCP server without complex middleware layers, or if you prefer established frameworks with larger community support and more comprehensive documentation. ### [mcp-server-tamplate](https://mcpcatalogs.com/en/server/gustta03-mcp-server-tamplate) by gustta03 · score 33 · ★0 TypeScript template for creating MCP servers with clean architecture and example tools. **When to choose:** Choose this template when starting a new MCP server project in TypeScript, especially if you value clean architecture and separation of concerns. **When NOT to choose:** Don't choose this if you need MCP servers in languages other than TypeScript or if you require more complex project structures beyond the template's scope. ### [mcp](https://mcpcatalogs.com/en/server/xp-forge-mcp) by xp-forge · score 33 · ★0 PHP implementation of Model Context Protocol for XP Framework with client and server components. **When to choose:** Choose this implementation if you're working within the XP Framework ecosystem or need to build MCP servers in PHP. **When NOT to choose:** Avoid if you're not using PHP or the XP Framework, as there are more established implementations in other languages. ### [mcp-client-manager-go](https://mcpcatalogs.com/en/server/vikashloomba-mcp-client-manager-go) by VikashLoomba · score 33 · ★3 A Go library that manages multiple MCP servers and provides a unified HTTP gateway for connecting to them. **When to choose:** Choose this package when building Go applications that need to coordinate multiple MCP servers or require a unified HTTP gateway to expose multiple MCP services through a single endpoint. **When NOT to choose:** Avoid this if you're working in environments without Go support or need language bindings for other programming languages beyond what the MCP protocol itself provides. ### [verbwire-mcp-server](https://mcpcatalogs.com/en/server/verbwire-verbwire-mcp-server) by verbwire · score 32 · ★2 A production-ready MCP server for blockchain operations through Verbwire API, supporting smart contracts, NFT minting, and IPFS storage. **When to choose:** Choose this MCP server when you need a comprehensive blockchain operations interface specifically for NFT workflows through Verbwire's API. **When NOT to choose:** Avoid this if you need support for blockchains not covered by Verbwire, or if you prefer a multi-provider solution rather than vendor lock-in to a single API provider. ### [mcp-graphql-tools](https://mcpcatalogs.com/en/server/saewoohan-mcp-graphql-tools) by saewoohan · score 32 · ★8 A Model Context Protocol server that enables AI assistants to interact with GraphQL APIs through standardized tools. **When to choose:** Choose this server when your AI assistant needs to interact with GraphQL APIs that have well-defined schemas, especially when you want to control query complexity and mutation permissions. **When NOT to choose:** Avoid this server if you need to interact with REST APIs instead of GraphQL, or if you require advanced caching mechanisms that aren't implemented in this basic server. ### [mcp-go-sdk](https://mcpcatalogs.com/en/server/setiabudiresearch-mcp-go-sdk) by SetiabudiResearch · score 32 · ★7 Go SDK for implementing MCP servers with tool execution, resource access, and prompt handling. **When to choose:** Choose this SDK when you need to build MCP servers in Go, especially for Go-centric development teams or when integrating with existing Go infrastructure. **When NOT to choose:** Avoid this if you need a more mature SDK with broader community support or if you're not already invested in the Go ecosystem. ### [First-MCP-Server-Project](https://mcpcatalogs.com/en/server/burakarslan0110-first-mcp-server-project) by burakarslan0110 · score 32 · ★0 A .NET 10 MCP server providing math and string processing tools for AI assistants. **When to choose:** Choose this if you're learning MCP server development in .NET or need a simple reference implementation for math and text operations. **When NOT to choose:** Not recommended for production use as it's likely a learning example with limited functionality and lack of advanced features like authentication or error handling. ### [server-nothing](https://mcpcatalogs.com/en/server/hesreallyhim-server-nothing) by hesreallyhim · score 32 · ★7 A minimalist test MCP server with no capabilities, designed for validating client/server setup without functionality. **When to choose:** When you're developing a new MCP client or server and need to test basic protocol connections before adding actual functionality. **When NOT to choose:** When you need any actual capabilities or functionality from your MCP server - this server intentionally provides no tools or resources. ### [Stellaris-Modding-MCP-Server](https://mcpcatalogs.com/en/server/kongyo2-stellaris-modding-mcp-server) by kongyo2 · score 32 · ★2 An MCP server for Stellaris modding that provides game version info, CWTools configs, and searchable documentation through SteamCMD and GitHub APIs. **When to choose:** Choose this if you're developing Stellaris mods and need real-time access to game documentation and version validation tools. **When NOT to choose:** Avoid if you're not working with Stellaris modding, as the tools are highly specialized for this specific game. ### [MCPWrapper](https://mcpcatalogs.com/en/server/jm12138-mcpwrapper) by jm12138 · score 31 · ★1 MCPWrapper is a Python tool that wraps any Python package as an MCP server without code modification. **When to choose:** Choose MCPWrapper when you need to quickly wrap existing Python libraries as MCP servers without code modifications. **When NOT to choose:** Avoid this if you need fine-grained control over the MCP implementation or require advanced authentication/authorization features. ### [motoko-mcp-example](https://mcpcatalogs.com/en/server/prometheus-protocol-motoko-mcp-example) by prometheus-protocol · score 31 · ★1 A Motoko template for creating Prometheus Protocol MCP servers with one-click deployment on ICP Ninja. **When to choose:** Choose this when developing MCP servers for the Internet Computer ecosystem, especially when you want quick prototyping without local setup or when planning to monetize through the Prometheus Protocol. **When NOT to choose:** Don't choose this if you need to deploy outside the Internet Computer ecosystem or if you prefer traditional development environments over the specialized ICP Ninja workflow. ### [infura-mcp](https://mcpcatalogs.com/en/server/deflang-infura-mcp) by deflang · score 31 · ★1 An MCP server for querying Ethereum blockchain data through Infura APIs using natural language. **When to choose:** Choose this MCP server when you need access to Ethereum blockchain data through natural language queries and are already using or willing to set up an Infura API key. **When NOT to choose:** Avoid this server if you need blockchain support beyond Ethereum, require robust write operations without manual enabling, or are concerned about potential vendor lock-in to Infura's infrastructure. ### [mcp-robot](https://mcpcatalogs.com/en/server/makarski-mcp-robot) by makarski · score 31 · ★1 Go-based MCP server SDK with type-safe tool definitions, fluent API, and automatic validation. **When to choose:** Teams developing MCP servers in Go who need type safety, fluent API, and a Go-native approach following standard Go patterns. **When NOT to choose:** Projects requiring full MCP protocol support (resources, prompts) or non-Go ecosystems, as this SDK currently only implements tools. ### [context-compose-mcp](https://mcpcatalogs.com/en/server/lazylagom-context-compose-mcp) by lazylagom · score 31 · ★1 Context Compose MCP server enables task-specific context composition for LLM interactions using YAML-based contexts. **When to choose:** Choose Context Compose when you need consistent, task-specific AI guidance across your development team, especially when working on complex projects with varying requirements. **When NOT to choose:** Don't choose Context Compose if you need simple, one-off AI interactions without the overhead of context management, or if you prefer all-in-one solutions rather than component-based approaches. ### [AIForwardDeveloperChallenge](https://mcpcatalogs.com/en/server/keidson299-aiforwarddeveloperchallenge) by keidson299 · score 31 · ★1 Python MCP server for Claude Desktop with code analysis and task management tools. **When to choose:** Choose this MCP server if you need basic code analysis and task management tools integrated with Claude Desktop, particularly for smaller development teams or individual developers. **When NOT to choose:** Avoid this if you need more advanced features, larger scale project management, or integration with development tools beyond Claude Desktop, as it has limited capabilities and a small community. ### [mcp-simple-demo](https://mcpcatalogs.com/en/server/arjunprabhulal-mcp-simple-demo) by arjunprabhulal · score 30 · ★4 A simple Python MCP server demonstrating basic tool implementation and client interactions. **When to choose:** Choose this for learning MCP basics or as a starting point for developing custom MCP tools in Python. **When NOT to choose:** Don't choose this for production use as it lacks authentication, error handling, and scalability features needed for real applications. ### [mcp-sdk-client-ssejs](https://mcpcatalogs.com/en/server/mybigday-mcp-sdk-client-ssejs) by mybigday · score 30 · ★0 A React Native-compatible client transport for MCP using SSE.js to support llm.rn integration. **When to choose:** Choose this when building React Native applications that need to connect to MCP servers and you're using llama.rn for LLM integration. **When NOT to choose:** Don't choose this if you're not working in React Native or don't need SSE-based transport for your MCP client implementation. ### [mcp-npm_docs-server](https://mcpcatalogs.com/en/server/bsmi021-mcp-npm-docs-server) by bsmi021 · score 30 · ★3 MCP server for fetching NPM package metadata and documentation with local SQLite caching. **When to choose:** Choose this MCP server when you need to quickly access NPM package documentation and metadata in AI workflows, especially with caching benefits for repeated queries. **When NOT to choose:** Don't choose this if you need real-time package data without caching, or if you require access to private NPM packages not available through npms.io. ### [mcp-server-hub-server](https://mcpcatalogs.com/en/server/bsmi021-mcp-server-hub-server) by bsmi021 · score 30 · ★3 A gateway that centralizes multiple MCP servers, allowing clients to access all tools through a single endpoint. **When to choose:** Choose this when you're using multiple MCP servers with different LLM clients and want to avoid configuring each server in every client. **When NOT to choose:** Don't choose this if you're only using a single MCP server or prefer direct connections between clients and servers without the additional abstraction layer. ### [mcp-agent](https://mcpcatalogs.com/en/server/ceerdecy-mcp-agent) by CeerDecy · score 30 · ★0 A Rust framework for integrating MCP tools with LLM clients, handling tool integration and client management. **When to choose:** Developers who want to build MCP tools with LLM integration and need a structured framework in Rust. **When NOT to choose:** If you're looking for an MCP server rather than a framework to build MCP tools, or if you prefer development in languages other than Rust. ### [ark-mcp](https://mcpcatalogs.com/en/server/vpopescu-ark-mcp) by vpopescu · score 30 · ★0 Enterprise MCP server supporting WebAssembly plugins as MCP tools. **When to choose:** Choose this server for enterprise environments needing WebAssembly plugin security and authentication via JWT tokens. **When NOT to choose:** Not suitable for simple use cases without plugin requirements or if you need extensive documentation. ### [mcp-python-executor](https://mcpcatalogs.com/en/server/bsmi021-mcp-python-executor) by bsmi021 · score 30 · ★3 An MCP server for executing Python code and managing packages with safety constraints. **When to choose:** Choose this server when you need to execute Python code in AI environments with controlled resource limits and package management. **When NOT to choose:** Avoid if you need to install system packages or if you require execution of Python scripts with long-running processes (>30s timeout). ### [jira-mcp-server](https://mcpcatalogs.com/en/server/parassolanki-jira-mcp-server) by ParasSolanki · score 30 · ★3 MCP server for Jira API integration with tools for listing projects, boards, sprints and issues. **When to choose:** Choose this server for basic Jira integration needs, especially if you primarily need to retrieve project, board, sprint, and issue information and create simple task issues. **When NOT to choose:** Don't choose this if you need advanced issue management features like updating/deleting issues, managing comments, or working with non-task issue types. ### [mcp-server](https://mcpcatalogs.com/en/server/agentico-dev-mcp-server) by agentico-dev · score 30 · ★3 A facade library that simplifies MCP server implementation by providing a base class and registration pattern. **When to choose:** Choose this facade when you want to quickly implement MCP servers without dealing with the complexity of the official SDK, especially for simple tools or team projects. **When NOT to choose:** Avoid this if you need advanced MCP features like resource management or prompts, as the facade may abstract away these capabilities. ### [MCP-Next.js](https://mcpcatalogs.com/en/server/sattarrasouli-mcp-next-js) by sattarrasouli · score 30 · ★0 A Next.js MCP server with TypeScript that exposes multiple simple tools like dice rolling and greetings via JSON-RPC API. **When to choose:** Developers looking for a simple Next.js-based MCP server implementation example with TypeScript to understand MCP architecture. **When NOT to choose:** Production environments requiring robust error handling, authentication, and scalability beyond demonstration purposes. ### [create-advanced-mcp-server](https://mcpcatalogs.com/en/server/bsmi021-create-advanced-mcp-server) by bsmi021 · score 29 · ★2 A CLI tool to scaffold MCP servers with web capabilities, TypeScript, and optional database integration. **When to choose:** Choose this tool when you need to build MCP servers with web APIs, real-time capabilities, or database persistence, and want a solid foundation with TypeScript. **When NOT to choose:** Don't choose this if you need a minimal MCP server without web or database features, or if you prefer JavaScript over TypeScript. ### [ATSPI-MCP-Server](https://mcpcatalogs.com/en/server/c-loftus-atspi-mcp-server) by C-Loftus · score 29 · ★2 An MCP server that exposes Linux accessibility tree information through ATSPI **When to choose:** Choose this when building AI agents that need to interact with or understand Linux GUI applications and accessibility state. **When NOT to choose:** Not suitable for Windows or macOS systems, or when you need comprehensive UI control beyond accessibility information. ### [apitomcp](https://mcpcatalogs.com/en/server/wrannaman-apitomcp) by wrannaman · score 27 · ★0 Converts any API into an MCP server using Swagger documentation. **When to choose:** Choose this for quick API-to-MCP conversion if you have Swagger documentation and need a simple integration point with Claude Desktop. **When NOT to choose:** Avoid this for production use due to lack of documentation, unclear API conversion process, and no evidence of robust error handling or security features. ### [mcp_tool_override_tester](https://mcpcatalogs.com/en/server/kibitzing-mcp-tool-override-tester) by kibitzing · score 25 · ★0 A Python MCP server demonstrating dynamic tool overriding across multiple servers. **When to choose:** When you need to test or demonstrate MCP tool overriding behavior across multiple servers in a development environment. **When NOT to choose:** For production use cases that require stable, non-conflicting tool implementations or robust error handling. --- ## ecommerce (24) ### [mcp-server-odoo](https://mcpcatalogs.com/en/server/ivnvxd-mcp-server-odoo) by ivnvxd · score 52 · ★282 MCP server enabling AI assistants to securely interact with Odoo ERP systems for data retrieval and manipulation. **When to choose:** Choose this MCP server if you need to integrate AI assistants with an existing Odoo ERP system for data retrieval, manipulation, and workflow automation. **When NOT to choose:** Avoid if you don't use Odoo, as this server is specifically designed for Odoo integration. Also avoid YOLO mode in production environments due to security risks. ### [shopify-mcp](https://mcpcatalogs.com/en/server/geli2001-shopify-mcp) by GeLi2001 · score 51 · ★209 Shopify MCP server with comprehensive tools for managing products, customers, orders and more through GraphQL API. **When to choose:** Choose this server if you need to integrate Shopify data with AI assistants or build custom workflows around e-commerce operations. **When NOT to choose:** Don't use this if you need access to other e-commerce platforms, or if you require read/write permissions beyond what Shopify's API provides. ### [meta-mcp](https://mcpcatalogs.com/en/server/brijr-meta-mcp) by brijr · score 50 · ★175 A comprehensive MCP server that enables AI assistants to interact with Facebook/Instagram advertising data through the Meta Marketing API. **When to choose:** Choose this if you need to integrate Meta advertising management into AI workflows and want comprehensive campaign control through natural language. **When NOT to choose:** Avoid if you need to manage non-Meta advertising platforms or require functionality beyond campaign creation, analytics, and audience management. ### [12306-mcp](https://mcpcatalogs.com/en/server/joooook-12306-mcp) by Joooook · score 49 · ★878 12306 MCP server provides ticket search functionality for Chinese railway booking. **When to choose:** Choose this server when you need to integrate real-time Chinese train ticket information into your AI application or workflow. **When NOT to choose:** Don't choose this server if you need access to non-Chinese railway systems or require ticket booking functionality (search only). ### [ebay-mcp](https://mcpcatalogs.com/en/server/yosefhayim-ebay-mcp) by YosefHayim · score 48 · ★63 eBay MCP Server provides AI assistants with access to 325 eBay API tools for inventory, orders, marketing, and analytics. **When to choose:** Choose this server for comprehensive eBay store management through AI assistants, particularly when you need high-volume API access with OAuth authentication. **When NOT to choose:** Avoid if you don't use eBay or need tools for other e-commerce platforms; this is specifically for eBay's ecosystem only. ### [ads-mcp](https://mcpcatalogs.com/en/server/amekala-ads-mcp) by amekala · score 48 · ★44 MCP server with 175+ tools for managing ad campaigns across Google Ads, Meta Ads, LinkedIn Ads, and TikTok Ads. **When to choose:** Choose Ads MCP when you need to manage ad campaigns across multiple platforms (Google, Meta, LinkedIn, TikTok) from a single interface and want strategy-aware execution that persists across sessions. **When NOT to choose:** Don't choose Ads MCP if you only work with a single ad platform or need extensive customization beyond what's provided by this commercial service. ### [mcp-server](https://mcpcatalogs.com/en/server/webflow-mcp-server) by webflow · score 48 · ★131 Webflow MCP server enables AI agents to interact with Webflow APIs through Model Context Protocol. **When to choose:** Choose this server if you're already using Webflow for website development and want AI assistance directly integrated into your workflow for content creation and site management. **When NOT to choose:** Avoid if you need write access to static content in the default locale or prefer a non-vendor-specific solution that doesn't lock you into Webflow's ecosystem. ### [meta-ads-mcp](https://mcpcatalogs.com/en/server/mikusnuz-meta-ads-mcp) by mikusnuz · score 47 · ★46 MCP server for Meta Marketing API v25.0 with 135 tools for Facebook & Instagram ad campaign management. **When to choose:** Choose this MCP server if you need comprehensive programmatic access to Meta's advertising platform for managing Facebook and Instagram ad campaigns through MCP tool calling. **When NOT to choose:** Avoid this if you need access to other advertising platforms besides Meta, or if you require offline ad management without an active connection to Meta's API. ### [mcd-mcp-server](https://mcpcatalogs.com/en/server/m-china-mcd-mcp-server) by M-China · score 47 · ★96 McDonald's China MCP Server provides food ordering, delivery, nutrition info, and loyalty program integration via the Model Context Protocol. **When to choose:** Choose this server if you need to integrate McDonald's China services into AI applications or workflows, especially for ordering, menu browsing, and customer loyalty features. **When NOT to choose:** Not suitable for international use as it's limited to mainland China and requires a Chinese mobile number for authentication. ### [rohlik-mcp](https://mcpcatalogs.com/en/server/tomaspavlin-rohlik-mcp) by tomaspavlin · score 47 · ★108 MCP server enabling AI assistants to shop groceries across Rohlik Group platforms in 6 European countries. **When to choose:** Choose this if you regularly shop through Rohlik Group platforms in Europe and want your AI assistant to help with smart grocery planning, personalized meal suggestions, and account management. **When NOT to choose:** Don't choose this if you need access to grocery platforms outside the Rohlik Group, as this server only works with their specific APIs and authentication systems. ### [konquest-meta-ads-mcp](https://mcpcatalogs.com/en/server/brandu-mos-konquest-meta-ads-mcp) by brandu-mos · score 46 · ★27 Production-ready MCP server with 57 Meta Ads tools and premium options for supervised campaign management. **When to choose:** Choose this MCP server if you need comprehensive Meta Ads management with safety gates, supervision, and optimization features already built for production environments. **When NOT to choose:** Avoid if you need write-only access without supervision, or if you're looking for a free solution with premium optimization features. ### [pancake-pos-mcp](https://mcpcatalogs.com/en/server/nguyennguyenit-pancake-pos-mcp) by nguyennguyenit · score 46 · ★28 MCP server for Pancake POS with 24 tools covering Vietnamese e-commerce operations. **When to choose:** Choose this MCP server if you're managing Vietnamese e-commerce operations through Pancake POS and need AI assistance for order management, inventory control, CRM, and multi-channel sales integration. **When NOT to choose:** Don't choose this if you're not using Pancake POS for your e-commerce operations, as it's specifically designed to integrate only with their API and offers no functionality for other POS systems. ### [google_ads_mcp](https://mcpcatalogs.com/en/server/google-marketing-solutions-google-ads-mcp) by google-marketing-solutions · score 46 · ★203 Google Ads MCP server enables LLMs to interact with Google Ads API via GAQL queries and ad management tools. **When to choose:** Choose this when you need AI assistance to interact with Google Ads for reporting, analysis, or campaign management, and you're already using Google Ads API. **When NOT to choose:** Don't choose this if you need production-ready tools as it's marked experimental, or if you require read/write access without additional configuration steps. ### [amazon-sp-mcp](https://mcpcatalogs.com/en/server/mansournorouzi-amazon-sp-mcp) by mansournorouzi · score 44 · ★29 Amazon SP-API MCP server connects Claude to Seller Central for sales, inventory, orders, and financial data using LWA OAuth authentication. **When to choose:** Choose this if you need Amazon Seller Central data access through Claude and prefer the simpler LWA-only authentication without AWS credentials. **When NOT to choose:** Not ideal if you need AWS-based SP-API authentication (for advanced features not covered by LWA) or if you require marketplace support beyond the currently covered regions. ### [woocommerce-mcp-server](https://mcpcatalogs.com/en/server/techspawn-woocommerce-mcp-server) by techspawn · score 44 · ★89 A comprehensive MCP server for WooCommerce integration with WordPress REST API, covering products, orders, customers, and store management. **When to choose:** Choose this MCP server if you need to integrate WooCommerce store management with AI workflows, especially if you're already using WordPress and WooCommerce for your e-commerce operations. **When NOT to choose:** Avoid this server if you're using a different e-commerce platform that doesn't run on WordPress, as it's specifically designed for WooCommerce integration. ### [mcp-hubspot](https://mcpcatalogs.com/en/server/baryhuang-mcp-hubspot) by baryhuang · score 44 · ★122 MCP server for HubSpot CRM integration with vector storage and caching to overcome API limitations. **When to choose:** Choose this if you need AI assistants to interact with your HubSpot CRM data, especially if you require semantic search capabilities and want to overcome API limitations through caching. **When NOT to choose:** Don't choose this if you need access to HubSpot features not covered by the available tools or if you require a different CRM platform integration. ### [chatbot_Shopify](https://mcpcatalogs.com/en/server/mobeen-dev-chatbot-shopify) by Mobeen-Dev · score 43 · ★25 Shopify chatbot with MCP integration that uses AI-powered conversations and agentic capabilities for seamless e-commerce experiences. **When to choose:** Choose this MCP server if you're running a Shopify store and want AI-powered customer support with seamless access to product data, order processing, and cart management capabilities. **When NOT to choose:** Don't choose this if you're not using Shopify, as it's tightly coupled to the Shopify ecosystem and requires Shopify API credentials. ### [mcp-server-amazon](https://mcpcatalogs.com/en/server/rigwild-mcp-server-amazon) by rigwild · score 38 · ★29 Unofficial Amazon MCP server enabling product search and purchase directly through Claude AI. **When to choose:** Choose this server if you want to shop on Amazon through AI assistants without switching between applications. **When NOT to choose:** Don't choose this for real Amazon purchases as orders are only for demonstration purposes, and it lacks official Amazon support. ### [eve-online-mcp](https://mcpcatalogs.com/en/server/kongyo2-eve-online-mcp) by kongyo2 · score 38 · ★7 MCP server for accessing EVE Online market data through ESI API with prices, orders, history, and structure market data. **When to choose:** Choose this server for EVE Online market analysis when you need comprehensive market data access through MCP with both public and authenticated endpoints. **When NOT to choose:** Don't choose this if you need real-time market data with sub-second updates, as it depends on ESI API which has inherent latency. ### [gumroad-mcp](https://mcpcatalogs.com/en/server/rmarescu-gumroad-mcp) by rmarescu · score 37 · ★22 An MCP server for Gumroad API, enabling AI assistants to manage products, sales, and offer codes. **When to choose:** Choose this if you're a Gumroad creator who wants to use AI to manage your store, analyze sales, and create promotional campaigns through natural language commands. **When NOT to choose:** Don't choose this if you need to manage multiple e-commerce platforms beyond Gumroad, as this server is specifically designed for Gumroad API only. ### [evetycoon-mcp-server](https://mcpcatalogs.com/en/server/kongyo2-evetycoon-mcp-server) by kongyo2 · score 36 · ★3 An MCP server for EVE Online market data, providing access to price statistics, order books, and historical information. **When to choose:** Choose this server if you need EVE Online market data integration for AI analysis or trading bots. **When NOT to choose:** Don't choose if you need real-time market data or require write access to the EVE database - this server only provides read-only access. ### [funpay-mcp](https://mcpcatalogs.com/en/server/dkeken-funpay-mcp) by DKeken · score 36 · ★0 MCP server for FunPay.com with profile management, lot operations, and batch cloning. **When to choose:** Choose this MCP server if you need to programmatically manage FunPay marketplace listings and are already using Bun as your runtime. **When NOT to choose:** Don't choose this if you need marketplace integration outside of FunPay or if you're not comfortable using Bun as a runtime dependency. ### [hotpepper-gourmet-mcp-server](https://mcpcatalogs.com/en/server/miyamo2-hotpepper-gourmet-mcp-server) by miyamo2 · score 35 · ★2 Unofficial MCP server for accessing Hotpepper Gourmet restaurant search APIs via Go implementation. **When to choose:** Choose this when you need to integrate Japanese restaurant and food location data into AI applications through the MCP protocol. **When NOT to choose:** Don't choose this if you need data outside Japan or prefer solutions with more comprehensive restaurant information beyond the Hotpepper Gourmet API. ### [wp-mcp](https://mcpcatalogs.com/en/server/adi-ty-wp-mcp) by Adi-ty · score 32 · ★2 WordPress MCP server enables AI agents to manage WordPress content through natural language commands. **When to choose:** Choose wp-mcp when you need AI-powered natural language control over your WordPress content management workflow. **When NOT to choose:** Avoid wp-mcp if you need multi-site management or advanced user permissions beyond what WordPress Application Passwords provide. --- ## education (4) ### [mcp-canvas-lms](https://mcpcatalogs.com/en/server/dmontgomery40-mcp-canvas-lms) by DMontgomery40 · score 49 · ★97 Comprehensive MCP server for Canvas LMS with 54+ tools for students, instructors, and administrators. **When to choose:** Choose this server if you work with Canvas LMS and need comprehensive tools for students, instructors, and administrators to interact with the platform through MCP. **When NOT to choose:** Avoid if you use a different LMS platform (like Moodle or Blackboard) or if you need highly specialized Canvas features not covered by the 54+ available tools. ### [aula-mcp](https://mcpcatalogs.com/en/server/casperjuel-aula-mcp) by Casperjuel · score 43 · ★22 MCP server for Denmark's Aula school platform with MitID authentication, enabling AI agents to access student profiles, calendar, messages and schedules. **When to choose:** Choose this if you're a parent or teacher in the Danish education system who needs AI access to Aula's school platform without using headless browsers. **When NOT to choose:** Don't choose this if you're outside Denmark, need write access to school data, or require a more widely supported educational platform integration. ### [brightspace-mcp-server](https://mcpcatalogs.com/en/server/rohanmuppa-brightspace-mcp-server) by RohanMuppa · score 43 · ★19 MCP server for Brightspace (D2L) that lets AI check grades, assignments, announcements and course content. **When to choose:** Choose this server if you're a student using D2L Brightspace and want to use AI assistants to quickly access and analyze your course information. **When NOT to choose:** Don't choose this if your school doesn't use D2L Brightspace, or if you need write access to your LMS (this server is read-only). ### [openedu-mcp](https://mcpcatalogs.com/en/server/cicatriiz-openedu-mcp) by Cicatriiz · score 38 · ★8 OpenEdu MCP Server provides educational resources and curriculum planning tools integrating OpenLibrary, Wikipedia, Dictionary, and arXiv APIs with grade-level filtering. **When to choose:** Choose this server for educational content discovery and curriculum planning, especially if you need age-appropriate filtering and alignment with educational standards. **When NOT to choose:** Don't choose this if you need broader general knowledge search without educational filtering, or if you require services not integrated with the current educational APIs. --- ## file-system (29) ### [filesystem](https://mcpcatalogs.com/en/server/modelcontextprotocol-servers-filesystem) by modelcontextprotocol · score 77 · ★85,748 · official A feature-rich MCP server for filesystem operations with dynamic directory access control. **When to choose:** Choose this when you need comprehensive file system operations with dynamic directory access control, especially when working with AI agents that need secure, scoped access to multiple directories. **When NOT to choose:** Avoid if you need fine-grained file permissions beyond directory-level access control, or if you're working with extremely sensitive systems where you need additional security layers beyond MCP's basic access model. ### [markdownify-mcp](https://mcpcatalogs.com/en/server/zcaceres-markdownify-mcp) by zcaceres · score 57 · ★2,686 MCP server converting various file types and web content to Markdown with comprehensive tool coverage. **When to choose:** Choose this server when you need to convert diverse file types and web content to Markdown format, especially if you're working with documents, research papers, or multimedia content that needs to be processed into readable text. **When NOT to choose:** Don't choose this if you need batch processing of extremely large files or if you require advanced editing capabilities beyond simple conversion, as it focuses primarily on content extraction and transformation. ### [pdf-reader-mcp](https://mcpcatalogs.com/en/server/sylphxai-pdf-reader-mcp) by SylphxAI · score 55 · ★711 Production-ready MCP server for high-performance PDF processing with 5-10x faster parallel processing and comprehensive test coverage. **When to choose:** Choose this when you need high-performance PDF processing for AI applications, especially with large documents where parallel processing makes a significant difference. **When NOT to choose:** Not ideal if you need to modify PDF content (this is read-only only) or require support for document formats other than PDF. ### [mcp-documentation-server](https://mcpcatalogs.com/en/server/andrea9293-mcp-documentation-server) by andrea9293 · score 52 · ★313 A TypeScript MCP server for local document management with semantic search using Orama vector database and optional Gemini AI integration. **When to choose:** Choose this server when you need local-first document management with semantic search capabilities, especially when working with sensitive data that cannot be stored in cloud services. **When NOT to choose:** Avoid this if you need multi-user authentication or enterprise-level features like collaborative editing, as it's designed primarily for single-user local use. ### [rust-mcp-filesystem](https://mcpcatalogs.com/en/server/rust-mcp-stack-rust-mcp-filesystem) by rust-mcp-stack · score 49 · ★149 A high-performance, asynchronous MCP server for filesystem operations written in Rust. **When to choose:** Choose this server when you need high-performance, asynchronous file system operations in a Rust-based environment with advanced glob pattern matching and ZIP archive support. **When NOT to choose:** Don't choose this if you need write access by default or require integration with languages other than Rust, as it's specifically designed for read-only operations by default. ### [ebook-mcp](https://mcpcatalogs.com/en/server/onebirdrocks-ebook-mcp) by onebirdrocks · score 49 · ★366 A Python MCP server for processing EPUB and PDF files with natural language queries. **When to choose:** Choose Ebook-MCP when you need to interact with your digital library through natural language and want to extract information from EPUB and PDF files using AI tools. **When NOT to choose:** Avoid this server if you need support for ebook formats other than EPUB and PDF, or if you require advanced OCR capabilities for image-based PDFs. ### [google-drive-mcp](https://mcpcatalogs.com/en/server/piotr-agier-google-drive-mcp) by piotr-agier · score 49 · ★151 A comprehensive MCP server for managing Google Drive files, Docs, Sheets, Slides, and Calendar with OAuth authentication. **When to choose:** Choose this MCP server if you need to integrate with Google Workspace services and want to manage files, documents, and calendars through MCP. **When NOT to choose:** Don't choose this if you need access to non-Google cloud storage services or prefer a self-hosted solution without OAuth dependencies. ### [kept](https://mcpcatalogs.com/en/server/egroup-labs-kept) by egroup-labs · score 48 · ★83 Kept is an MCP server that manages AI conversations as searchable Markdown files in a local vault. **When to choose:** Choose Kept when you want to own your AI conversation history as local files with powerful search and need MCP integration for accessing that archive with agents. **When NOT to choose:** Don't choose Kept if you prefer cloud-based storage or need multi-device sync as it's entirely local with no cloud component. ### [hwpx-mcp-server](https://mcpcatalogs.com/en/server/airmang-hwpx-mcp-server) by airmang · score 48 · ★57 MCP server for AI agents to read, edit, inspect, and validate local HWPX documents without Hangul word processor. **When to choose:** Choose this server when working with Korean HWPX documents and need AI agents to read, edit, or validate them without installing Hangul word processor. **When NOT to choose:** Don't choose if you need to work with binary .hwp files, require collaborative editing features, or need authentication/permission controls for document access. ### [mcp-server-synology](https://mcpcatalogs.com/en/server/atom2ueki-mcp-server-synology) by atom2ueki · score 47 · ★99 MCP server for Synology NAS enabling AI assistants to manage files, downloads, and system operations through secure API integration. **When to choose:** Choose this MCP server if you use a Synology NAS and want AI assistants to manage files, downloads, and monitor system health through secure API integration. **When NOT to choose:** Avoid if you use non-Synology NAS devices, need write capabilities beyond what's provided, or require enhanced security features like 2FA support. ### [mcp-workspace](https://mcpcatalogs.com/en/server/marcusjellinghaus-mcp-workspace) by MarcusJellinghaus · score 45 · ★47 Secure MCP server providing file system, Git, and GitHub tools for AI assistants in a sandboxed project directory. **When to choose:** Choose this when you need secure, sandboxed file system access for AI assistants in local development projects with support for multiple reference projects. **When NOT to choose:** Avoid if you need cloud-based file access, broader system permissions beyond project directories, or integration with services other than GitHub. ### [markitdown_mcp_server](https://mcpcatalogs.com/en/server/korigamik-markitdown-mcp-server) by KorigamiK · score 45 · ★71 MCP server converting multiple file formats to Markdown using MarkItDown utility. **When to choose:** Choose this when you need to convert multiple document formats to Markdown for content analysis or research, especially when working with heterogeneous file collections. **When NOT to choose:** Not suitable for applications requiring high-fidelity document preservation with complex formatting, as the output is limited to Markdown. ### [mcp-tool-kit](https://mcpcatalogs.com/en/server/getfounded-mcp-tool-kit) by getfounded · score 44 · ★107 MCP Tool Kit provides a Python framework with 120+ tools for building vertical AI agents with automatic discovery. **When to choose:** Choose MCP Tool Kit when you need a comprehensive set of pre-built tools for file system operations, web search, and browser automation, especially if you're building vertical AI agents. **When NOT to choose:** Avoid this if you need lightweight solution without Docker dependency or want maximum control over the underlying MCP implementation without the abstraction layer. ### [mcp-file-context-server](https://mcpcatalogs.com/en/server/bsmi021-mcp-file-context-server) by bsmi021 · score 43 · ★36 A robust MCP server providing file system context to LLMs with advanced caching and code analysis capabilities. **When to choose:** Choose this MCP server when you need comprehensive file system access with caching for LLMs, particularly for code analysis and real-time file monitoring in development environments. **When NOT to choose:** Don't choose this if you need write access to files (it's read-only) or if you require authentication for accessing restricted file systems. ### [aiquila](https://mcpcatalogs.com/en/server/elgorro-aiquila) by elgorro · score 43 · ★17 AIquila is an MCP server connecting Claude AI to Nextcloud for file and data management. **When to choose:** Choose this if you're using Nextcloud and want to integrate Claude AI for file management without data export. **When NOT to choose:** Don't choose if you need AI capabilities outside Nextcloud ecosystem or if you're not comfortable with AGPL-3.0/MIT licensing. ### [obsidian-mcp-server](https://mcpcatalogs.com/en/server/smith-and-web-obsidian-mcp-server) by smith-and-web · score 42 · ★15 MCP server enabling AI assistants to manage Obsidian vaults with comprehensive note, tag, and search operations. **When to choose:** Choose this server if you use Obsidian as your knowledge base and want AI assistants to directly interact with your vault through natural language commands. **When NOT to choose:** Don't choose this if you need access to other note-taking applications like Notion or Roam Research, as it's specifically designed for Obsidian vaults. ### [mcp-obsidian](https://mcpcatalogs.com/en/server/piotr1215-mcp-obsidian) by Piotr1215 · score 42 · ★22 An MCP server for secure, direct access to Obsidian vaults without the Obsidian app. **When to choose:** Choose this server when you need direct file system access to Obsidian vaults without requiring the Obsidian app to be running, especially for automation workflows or integration with vim-based setups like obsidian.nvim. **When NOT to choose:** Don't choose this if you need remote access to Obsidian vaults, as this server only works with local file systems, or if you require Obsidian plugin integration that's only available through the REST API. ### [filesystem-mcp-server](https://mcpcatalogs.com/en/server/cyanheads-filesystem-mcp-server) by cyanheads · score 41 · ★37 A robust TypeScript MCP server enabling AI agents to securely interact with the local filesystem via STDIO or HTTP transports. **When to choose:** Choose this when you need comprehensive file system operations for your AI agent, especially when working in restricted environments where sandboxed file access is required. **When NOT to choose:** Avoid if you need cloud storage integration or if your application requires write access to multiple cloud providers as this server only handles local filesystem operations. ### [mcp-file-operations-server](https://mcpcatalogs.com/en/server/bsmi021-mcp-file-operations-server) by bsmi021 · score 40 · ★21 A comprehensive MCP server with streaming support, file operations, change tracking, and HTTP interface for enhanced file management. **When to choose:** Choose this server when you need comprehensive file operations with advanced features like change tracking, file watching, and progress reporting for both local and remote access scenarios. **When NOT to choose:** Avoid this if you need only basic file operations without the overhead of streaming and tracking features, or if you require transport protocols beyond stdio and HTTP. ### [mcp-windows-automation](https://mcpcatalogs.com/en/server/mukul975-mcp-windows-automation) by mukul975 · score 40 · ★22 AI-powered Windows automation server using MCP with 80+ tools for system control, file operations, and more. **When to choose:** Choose this server if you need Windows-specific automation with AI assistants and require comprehensive control over system applications through natural language. **When NOT to choose:** Avoid this if you need cross-platform automation support or require high-level enterprise security features, as it's Windows-specific with basic security controls. ### [mcp-server-text-editor](https://mcpcatalogs.com/en/server/bhouston-mcp-server-text-editor) by bhouston · score 38 · ★36 An MCP server implementation of Claude's built-in text editor tool for viewing, editing, and creating text files. **When to choose:** Choose this server when you need to implement file editing capabilities for Claude in environments where the built-in editor isn't available, or when you need more control over the file editing process. **When NOT to choose:** Don't choose this if you need advanced file operations beyond text editing, such as binary file manipulation or complex version control workflows. ### [Convert-Markdown-PDF-MCP](https://mcpcatalogs.com/en/server/seanivore-convert-markdown-pdf-mcp) by seanivore · score 38 · ★15 MCP server that converts Markdown to PDF with VS Code styling using Python's ReportLab. **When to choose:** Choose this server if you need to convert Markdown documents to professionally styled PDFs and want to integrate this functionality into an MCP-enabled workflow. **When NOT to choose:** Don't choose this if you need advanced PDF customization options, batch processing of multiple files, or conversion of formats other than Markdown. ### [LocalFilesOrganizer-MCP](https://mcpcatalogs.com/en/server/tobiasarg-localfilesorganizer-mcp) by TobiasArg · score 36 · ★1 MCP server with tools to safely organize, read, write, and analyze local files with built-in protections. **When to choose:** Choose this server when you need a dedicated MCP interface for safe local file operations with specific organizational needs and security constraints. **When NOT to choose:** Avoid this if you need broader file system access outside the allowed paths, or require more advanced metadata or content analysis capabilities not provided here. ### [copilot-memory-store](https://mcpcatalogs.com/en/server/tokio17-copilot-memory-store) by Tokio17 · score 34 · ★1 A tool that stores and manages local JSON data for context engineering with GitHub Copilot and MCP clients. **When to choose:** Choose this tool when you need simple local JSON storage for GitHub Copilot context and want CLI, MCP server, and VS Code integration options. **When NOT to choose:** Don't choose this if you need enterprise features, multi-user support, or cloud-based storage solutions. ### [claude-memory-mcp](https://mcpcatalogs.com/en/server/tanaka-kun-dev-claude-memory-mcp) by tanaka-kun-dev · score 33 · ★0 MCP server for persistent file-based memory management with tools to read, search, and update Claude's memory files. **When to choose:** Choose this server if you're already using Claude's memory feature and need to manage it across multiple MCP clients, or if you want programmatic access to Claude's persistent memories. **When NOT to choose:** Don't choose this if you need memory storage outside of Claude's ecosystem, or if you require more sophisticated memory management with relationships or structured data. ### [MCP-PDF-Extractor-server](https://mcpcatalogs.com/en/server/rayenmalouche-mcp-pdf-extractor-server) by RayenMalouche · score 33 · ★0 Java-based MCP server using Apache Tika to extract content and metadata from PDFs, DOCX and other documents. **When to choose:** Choose this server for local document processing workflows where you need to extract content and metadata without exposing documents to external services. **When NOT to choose:** Avoid if you need cloud-based processing or have already established infrastructure in other languages like Python. ### [Verzeichnis-Server-MCP](https://mcpcatalogs.com/en/server/stephanj79-verzeichnis-server-mcp) by stephanj79 · score 32 · ★2 A C# MCP server providing secure, controlled file system operations for AI agents. **When to choose:** Choose this server when you need a .NET-based solution for secure file system access in AI workflows, especially if you're already using N8N. **When NOT to choose:** Don't choose this server if you need non-.NET environments, as it's limited to C#/.NET runtime and requires Windows for optimal performance. ### [go-mcp-server-mds](https://mcpcatalogs.com/en/server/warashi-go-mcp-server-mds) by Warashi · score 28 · ★1 Go-based MCP server serving markdown files with frontmatter support from filesystem. **When to choose:** Choose this server when you need to expose local markdown files with frontmatter to AI agents through MCP, particularly when working in Go environments. **When NOT to choose:** Don't choose this if you need to serve markdown from remote locations, databases, or content management systems beyond local filesystems. ### [McpToolsSdkBased](https://mcpcatalogs.com/en/server/srogmann-mcptoolssdkbased) by srogmann · score 27 · ★0 Java-based MCP server SDK providing file search, read, and edit tools. **When to choose:** Choose this server when you need a Java-based implementation for MCP file operations or as a starting point for developing custom MCP tools in Java. **When NOT to choose:** Avoid if you need advanced file operations, authentication, or a production-ready solution with comprehensive documentation and maintenance. --- ## finance (84) ### [akshare-one-mcp](https://mcpcatalogs.com/en/server/zwldarren-akshare-one-mcp) by zwldarren · score 75 · ★172 · 13,664 uses/30d MCP server providing comprehensive Chinese stock market data through akshare-one API. **When to choose:** Choose this server if you need access to Chinese stock market data within an MCP-enabled environment, especially when working with financial analysis tools. **When NOT to choose:** Don't choose this if you need data for markets outside of China, as it's specifically designed for Chinese stock market data only. ### [QuantDinger](https://mcpcatalogs.com/en/server/brokermr810-quantdinger) by brokermr810 · score 58 · ★5,392 QuantDinger is an AI-powered quantitative trading platform with MCP server integration for market research and trading operations. **When to choose:** Choose QuantDinger if you need a self-hosted quantitative trading platform with AI integration, particularly when you want full control over data and execution while leveraging AI agents for strategy development. **When NOT to choose:** Avoid QuantDinger if you're looking for a simple trading bot without the complex architecture, or if you need mobile-first capabilities without building the app yourself. ### [tradingview-mcp](https://mcpcatalogs.com/en/server/atilaahmettaner-tradingview-mcp) by atilaahmettaner · score 57 · ★2,698 AI trading MCP server with real-time stock/crypto screening, technical analysis, and native Claude integration. **When to choose:** Choose this MCP server if you need AI-powered trading tools with backtesting capabilities and technical analysis for multiple asset classes, especially if you want to avoid the complexity of setting up multiple data sources and API keys. **When NOT to choose:** Avoid this if you need real-time trading execution, require futures/options trading features, or prefer a cloud-based solution rather than self-hosting. ### [tradememory-protocol](https://mcpcatalogs.com/en/server/mnemox-ai-tradememory-protocol) by mnemox-ai · score 54 · ★906 TradeMemory Protocol is an MCP server providing decision audit trails and persistent memory for AI trading agents with outcome-weighted recall and SHA-256 tamper detection. **When to choose:** Choose this when building regulated trading AI systems that require persistent memory, audit trails, and behavioral analysis across trading sessions. **When NOT to choose:** Not suitable if you need a general-purpose memory solution or if you're building simple trading bots without compliance requirements. ### [google-meta-ads-ga4-mcp](https://mcpcatalogs.com/en/server/irinabuht12-oss-google-meta-ads-ga4-mcp) by irinabuht12-oss · score 54 · ★987 Unified MCP server for Google Ads, Meta Ads & GA4 with 250+ tools for cross-platform campaign management and analytics. **When to choose:** Choose this if you need to manage advertising campaigns across multiple platforms (Google Ads, Meta Ads, GA4) from a single MCP endpoint and want full read/write access rather than limited read-only access. **When NOT to choose:** Don't choose this if you need to work offline or if you're concerned about vendor lock-in with the Ryze AI hosting platform. ### [maverick-mcp](https://mcpcatalogs.com/en/server/wshobson-maverick-mcp) by wshobson · score 53 · ★549 A comprehensive MCP server for financial analysis, technical indicators, and portfolio optimization. **When to choose:** Choose MaverickMCP if you're an individual trader or investor looking for professional-grade stock analysis tools directly in Claude Desktop, with S&P 500 coverage and technical indicators. **When NOT to choose:** Avoid this if you need real-time data with minimal latency, as it relies on Tiingo API which may have rate limits, or if you need analysis outside of S&P 500 stocks. ### [FinanceMCP](https://mcpcatalogs.com/en/server/guangxiangdebizi-financemcp) by guangxiangdebizi · score 53 · ★574 Comprehensive financial data MCP server providing real-time stock, crypto, macro-economic data and technical indicators through Tushare and Binance APIs. **When to choose:** Choose FinanceMCP when you need comprehensive financial data access for AI assistants, especially for Chinese markets, technical analysis, or when working with Claude Desktop. **When NOT to choose:** Avoid FinanceMCP if you need data for markets outside its supported scope, require real-time data with minimal latency (cloud service has limitations), or prefer self-hosted solutions without external API dependencies. ### [alpaca-mcp-server](https://mcpcatalogs.com/en/server/alpacahq-alpaca-mcp-server) by alpacahq · score 53 · ★736 Alpaca's official MCP server enables natural language trading of stocks, ETFs, crypto, and options through AI assistants. **When to choose:** Choose this when you need AI-powered trading capabilities directly through LLM tools and want comprehensive market data for stocks, crypto, and options. **When NOT to choose:** Not suitable for users who need multi-brokerage support or require more sophisticated algorithmic trading tools beyond basic order types. ### [agent-trade-kit](https://mcpcatalogs.com/en/server/okx-agent-trade-kit) by okx · score 52 · ★297 OKX trading MCP server with 150+ tools for spot, swap, futures, options, bots and portfolio management. **When to choose:** Choose this server if you need comprehensive trading tools for OKX exchange and want to integrate them directly into AI agents like Claude or Cursor. **When NOT to choose:** Don't choose this if you need support for exchanges other than OKX, or if you prefer cloud-based trading solutions with remote access. ### [bank-api](https://mcpcatalogs.com/en/server/erwinkramer-bank-api) by erwinkramer · score 52 · ★833 A banking API reference implementation with MCP server for AI integration. **When to choose:** Choose this MCP server for banking API integrations when you need a reference implementation with strong security compliance and authentication patterns. **When NOT to choose:** Avoid for production banking systems as this is a reference implementation; also not suitable if you need non-financial API operations. ### [ai-trader](https://mcpcatalogs.com/en/server/whchien-ai-trader) by whchien · score 52 · ★694 AI trader MCP server allows LLMs to run backtests, fetch market data, and analyze trading strategies across stocks, crypto, and forex. **When to choose:** Choose this if you're developing algorithmic trading strategies and want to integrate LLM assistants directly into your backtesting workflow through MCP. **When NOT to choose:** Don't choose this if you need a different license (it's GPL-3.0) or if you require real-time trading capabilities (it's only for backtesting). ### [mcp_massive](https://mcpcatalogs.com/en/server/massive-com-mcp-massive) by massive-com · score 52 · ★332 An MCP server providing access to Massive.com financial market data via search, call, and query tools with in-memory SQLite storage. **When to choose:** Choose this server when you need comprehensive financial market data through an AI-friendly interface with built-in analytical functions. **When NOT to choose:** Don't choose this if you need access to other financial data providers beyond Massive.com, as this server is limited to their API ecosystem. ### [blockrun-mcp](https://mcpcatalogs.com/en/server/blockrunai-blockrun-mcp) by BlockRunAI · score 52 · ★464 Real-time data MCP server providing markets, research, X/Twitter and crypto data with pay-per-call billing. **When to choose:** Choose BlockRun when you need unified access to multiple AI models and real-time data sources without managing individual API keys and subscriptions. **When NOT to choose:** Avoid BlockRun if you require free access to data, as all services require micropayments, or if you need highly specialized APIs not available through their platform. ### [sec-edgar-mcp](https://mcpcatalogs.com/en/server/stefanoamorelli-sec-edgar-mcp) by stefanoamorelli · score 51 · ★266 MCP server providing AI assistants with precise access to SEC EDGAR financial filings and company data. **When to choose:** Choose SEC EDGAR MCP when you need precise, numeric financial data from SEC filings with direct URLs for verification in AI applications. **When NOT to choose:** Avoid using this server if you need commercial licensing or authentication capabilities, as it uses AGPL-3.0 license and has no authentication model. ### [razorpay-mcp-server](https://mcpcatalogs.com/en/server/razorpay-razorpay-mcp-server) by razorpay · score 51 · ★220 Official Razorpay MCP Server for payment processing with remote/local setup options. **When to choose:** Choose this if you're building AI applications that need to interact with Razorpay's payment ecosystem, especially when you want zero infrastructure maintenance. **When NOT to choose:** Don't choose this if you need to work with payment providers other than Razorpay, as this is vendor-specific with no support for alternative payment gateways. ### [mcp-aktools](https://mcpcatalogs.com/en/server/aahl-mcp-aktools) by aahl · score 50 · ★381 MCP server providing stock and cryptocurrency data query and analysis tools. **When to choose:** Choose this server when you need comprehensive financial data access for stocks and cryptocurrencies within your AI agent or workflow tool. **When NOT to choose:** Don't choose this if you need advanced trading functionality, real-time order placement, or extensive backtesting capabilities beyond data retrieval. ### [Financial-Modeling-Prep-MCP-Server](https://mcpcatalogs.com/en/server/imbenrabi-financial-modeling-prep-mcp-server) by imbenrabi · score 50 · ★131 MCP server providing 250+ financial data tools for stocks, crypto, economics, and market analysis. **When to choose:** Choose this server when you need comprehensive financial data for AI applications, especially if you're already using the Financial Modeling Prep API or want to access 250+ financial tools through MCP. **When NOT to choose:** Avoid if you need write access to financial data (this is read-only only) or if you're concerned about single-vendor lock-in to the Financial Modeling Prep service. ### [monarch-mcp-server](https://mcpcatalogs.com/en/server/robcerda-monarch-mcp-server) by robcerda · score 50 · ★216 MCP server for Monarch Money personal finance management with extensive transaction and account tools. **When to choose:** Choose this if you're already using Monarch Money for personal finance and want to leverage Claude's AI capabilities for financial analysis and management. **When NOT to choose:** Don't choose this if you're not a Monarch Money user or if you need read/write access to other financial platforms not supported by this server. ### [web3-research-mcp](https://mcpcatalogs.com/en/server/aaronjmars-web3-research-mcp) by aaronjmars · score 49 · ★153 A local MCP server for comprehensive cryptocurrency research across multiple sources. **When to choose:** Choose this server when you need comprehensive cryptocurrency research capabilities without recurring API costs and want structured analysis of tokens and DeFi protocols. **When NOT to choose:** Avoid if you need real-time trading data or require API access to centralized exchanges, as this focuses on research rather than trading functionality. ### [trading_skills](https://mcpcatalogs.com/en/server/staskh-trading-skills) by staskh · score 49 · ★208 MCP server providing 32+ trading analysis tools for options traders and market analysis. **When to choose:** Choose this if you're an options trader or investor who wants conversational access to market data, technical analysis, and portfolio management tools through Claude Desktop without switching between multiple applications. **When NOT to choose:** Avoid this if you need write capabilities to your broker account (most IB features are read-only) or if you don't primarily trade options as the server is specialized for options trading strategies. ### [octagon-mcp-server](https://mcpcatalogs.com/en/server/octagonai-octagon-mcp-server) by OctagonAI · score 49 · ★122 A financial research MCP server providing market intelligence through API integration with Claude Desktop. **When to choose:** Choose Octagon MCP for specialized financial research, particularly when analyzing public filings, earnings calls, or private market data through AI-powered agents. **When NOT to choose:** Avoid if you need free access without API key requirements, or if you require tools outside of financial market research capabilities. ### [yahoo-finance-mcp](https://mcpcatalogs.com/en/server/alex2yang97-yahoo-finance-mcp) by Alex2Yang97 · score 49 · ★286 Comprehensive Yahoo Finance MCP server providing stock data, financial statements, options, and market news. **When to choose:** Choose this MCP server if you need comprehensive financial data from Yahoo Finance through MCP for stock analysis, market research, or investment research purposes. **When NOT to choose:** Not ideal if you need real-time data (as Yahoo Finance has some delay), require data sources beyond Yahoo Finance, or need authentication/enterprise features. ### [yfinance-mcp](https://mcpcatalogs.com/en/server/narumiruna-yfinance-mcp) by narumiruna · score 49 · ★135 Yahoo Finance MCP server providing stock data, financial statements, news, charts, and options data. **When to choose:** When you need comprehensive financial market data integration with AI assistants, particularly for stock analysis, news retrieval, and technical chart generation. **When NOT to choose:** When you need real-time trading capabilities or data from financial sources beyond Yahoo Finance, as this server only provides read-only access to Yahoo data. ### [fred-mcp-server](https://mcpcatalogs.com/en/server/stefanoamorelli-fred-mcp-server) by stefanoamorelli · score 49 · ★92 FRED MCP Server provides universal access to Federal Reserve Economic Data through three powerful tools. **When to choose:** Choose this server when you need access to authoritative US economic data for research, financial analysis, or educational purposes without building custom API integrations. **When NOT to choose:** Avoid this if you need non-US economic data or if you require write access to the data source as this is a read-only API wrapper. ### [evm-mcp-server](https://mcpcatalogs.com/en/server/mcpdotdirect-evm-mcp-server) by mcpdotdirect · score 49 · ★376 Comprehensive MCP server providing blockchain interaction tools across 60+ EVM networks with 22 tools and AI-guided prompts. **When to choose:** Choose this server when your AI agents need to interact with blockchain networks, transfer tokens, query balances, or work with smart contracts across multiple EVM chains. **When NOT to choose:** Don't choose this if you need blockchain functionality for non-EVM chains like Bitcoin or Solana, or if you require enterprise-grade security features like multi-signature wallets. ### [interactive-brokers-mcp](https://mcpcatalogs.com/en/server/code-rabi-interactive-brokers-mcp) by code-rabi · score 48 · ★131 A comprehensive MCP server for Interactive Brokers trading with market data, position tracking, and order placement capabilities. **When to choose:** Choose this MCP server if you need to integrate AI assistants with Interactive Brokers for trading operations and financial data retrieval, especially when you prefer using standard npm commands without additional installations. **When NOT to choose:** Don't choose this if you're using a different broker platform, need write access to trading accounts beyond what IB offers, or require production-level stability for mission-critical trading operations. ### [stock-mcp](https://mcpcatalogs.com/en/server/huweihua123-stock-mcp) by huweihua123 · score 48 · ★146 A professional financial market data MCP server supporting A-shares/US stocks/crypto with HTTP API for AI agents. **When to choose:** Choose stock-mcp when you need a unified financial data API for AI agents that covers both Chinese and international markets with multiple data sources and technical analysis capabilities. **When NOT to choose:** Don't choose this if you need specialized derivatives trading support or if you require write capabilities to market data (this is a read-only service). ### [us-gov-open-data-mcp](https://mcpcatalogs.com/en/server/lzinga-us-gov-open-data-mcp) by lzinga · score 48 · ★98 MCP server with 300+ tools for 40+ US government APIs including Treasury, FRED, Congress, FDA, CDC, and FEC. **When to choose:** Choose this server when you need comprehensive access to US government data for economic analysis, health research, or legislative tracking, especially when working with Claude Desktop, VS Code Copilot, or Cursor. **When NOT to choose:** Avoid this if you need access to non-US government data, as it focuses exclusively on US federal sources, or if you require write capabilities to government systems (this is read-only only). ### [mcp-google-ads](https://mcpcatalogs.com/en/server/cohnen-mcp-google-ads) by cohnen · score 48 · ★609 A Python MCP server that bridges Google Ads with Claude AI for natural language advertising data analysis and management. **When to choose:** Choose this if you're managing Google Ads campaigns and want to analyze performance data through natural language conversations with AI assistants. **When NOT to choose:** Avoid this if you need to manage advertising campaigns on platforms other than Google Ads, or if you prefer not to set up API authentication for Google services. ### [okx-agent-trade-kit](https://mcpcatalogs.com/en/server/placenl2026-okx-agent-trade-kit) by PlaceNL2026 · score 48 · ★86 OKX trading MCP server with 145 tools across 11 modules for crypto spot/futures/swap trading and account management. **When to choose:** Choose this MCP server if you're already using OKX for trading and want AI-powered access to your account without switching between your AI client and exchange interface. **When NOT to choose:** Don't choose this if you primarily trade on exchanges other than OKX, as it's specifically designed for the OKX platform only. ### [mcp-metatrader5-server](https://mcpcatalogs.com/en/server/qoyyuum-mcp-metatrader5-server) by Qoyyuum · score 48 · ★133 MCP server for MetaTrader 5 trading platform that provides AI assistants with market data and trading capabilities. **When to choose:** Choose this server if you need AI integration with MetaTrader 5 for automated trading or market analysis, especially if you're already using Python for development. **When NOT to choose:** Don't choose this if you're not using MetaTrader 5, as it's specifically designed for this platform and requires Windows installation. ### [TWSEMCPServer](https://mcpcatalogs.com/en/server/twjackysu-twsemcpserver) by twjackysu · score 48 · ★90 Comprehensive MCP server for Taiwan stock market data with real-time quotes and financial analysis. **When to choose:** Choose this server if you need comprehensive Taiwan stock market data with real-time quotes and financial analysis tools integrated through MCP. **When NOT to choose:** Don't choose this if you need data for markets outside Taiwan, or if you require write access to financial systems (this is a read-only data source). ### [lunchmoney-mcp](https://mcpcatalogs.com/en/server/akutishevsky-lunchmoney-mcp) by akutishevsky · score 48 · ★70 TypeScript MCP server for LunchMoney personal finance API with comprehensive transaction, budget, and account management. **When to choose:** Choose this if you use LunchMoney for personal finance management and want AI assistants to interact with your financial data through MCP. **When NOT to choose:** Don't choose this if you don't use LunchMoney or need access to other personal finance platforms that don't integrate with it. ### [ccxt-mcp](https://mcpcatalogs.com/en/server/lazy-dinosaur-ccxt-mcp) by lazy-dinosaur · score 47 · ★85 CCXT MCP Server enables AI models to interact with 100+ cryptocurrency exchanges through MCP protocol. **When to choose:** Choose this MCP server if you need AI assistants to directly interact with cryptocurrency exchanges for automated trading or portfolio analysis across multiple platforms. **When NOT to choose:** Avoid using this server if you need exchange features not covered by CCXT library or if you have strict security requirements preventing external API access. ### [sp500-mcp-server](https://mcpcatalogs.com/en/server/zhensherlock-sp500-mcp-server) by zhensherlock · score 47 · ★96 A well-documented MCP server providing S&P 500 company data tools through a Next.js web interface with Supabase backend. **When to choose:** Choose this MCP server if you need access to S&P 500 company data for financial analysis or AI applications, especially if you're already using or comfortable with Supabase as a data source. **When NOT to choose:** Don't choose this if you need data outside of S&P 500 companies, require real-time market data feeds, or prefer a self-hosted solution without Supabase dependencies. ### [world-intel-mcp](https://mcpcatalogs.com/en/server/marc-shade-world-intel-mcp) by marc-shade · score 47 · ★26 A comprehensive MCP server with 110+ tools for real-time global intelligence across markets, geopolitics, military, climate, and other domains. **When to choose:** Choose this MCP server for comprehensive real-time global intelligence across multiple domains when you need deep market insights, geopolitical monitoring, or environmental data for AI agents. **When NOT to choose:** Avoid this if you need domain-specific deep expertise in a single area, or if you require offline capabilities as all data relies on external public APIs. ### [openaccountants](https://mcpcatalogs.com/en/server/openaccountants-openaccountants) by openaccountants · score 46 · ★66 Open-source tax skills for AI across 134 countries, accessible via MCP or manual upload. **When to choose:** Choose OpenAccountants when you need automated tax computation skills for multiple countries, especially if you're already using Claude Desktop or Cursor with MCP support. **When NOT to choose:** Don't use OpenAccountants for critical tax filing without professional verification, as many skills are Q3 or below. It's not suitable for jurisdictions with incomplete coverage or where you need absolute certainty in automated tax calculations. ### [openfinclaw-cli](https://mcpcatalogs.com/en/server/mirror29-openfinclaw-cli) by mirror29 · score 46 · ★63 OpenFinClaw is an MCP server that provides AI-powered quant trading research, strategy generation, and backtesting capabilities across multiple AI platforms. **When to choose:** Choose OpenFinClaw when you need AI-powered quantitative trading research and strategy testing capabilities integrated directly into your AI workflow. **When NOT to choose:** Avoid if you need direct market data integration or real trading execution capabilities, as this focuses on research and backtesting only. ### [formanator](https://mcpcatalogs.com/en/server/timrogers-formanator) by timrogers · score 46 · ★84 Command-line tool and MCP server for submitting and managing Forma benefit claims with receipt analysis. **When to choose:** Choose Formanator if you're already using Forma for benefits management and want to integrate claim submission into your AI workflows or automate receipt processing. **When NOT to choose:** Don't choose Formanator if you don't use Forma, as it's specifically designed for this platform's benefits ecosystem. ### [dexpaprika-mcp](https://mcpcatalogs.com/en/server/coinpaprika-dexpaprika-mcp) by coinpaprika · score 46 · ★40 A comprehensive MCP server for accessing real-time crypto and DEX data across multiple blockchains with 14 specialized tools. **When to choose:** Choose DexPaprika MCP when you need access to real-time and historical DEX data without API key management, especially for cryptocurrency analysis and portfolio tracking applications. **When NOT to choose:** Don't choose this if you need more than 10,000 API requests per day, require write access to blockchain data, or need extensive historical data beyond 1 year for OHLCV queries. ### [mercury-invoicing-mcp](https://mcpcatalogs.com/en/server/klodr-mercury-invoicing-mcp) by klodr · score 45 · ★20 MCP server providing full access to Mercury Banking APIs with advanced invoicing features. **When to choose:** Choose this MCP when you need full Mercury banking access including invoicing automation and internal transfers between your own accounts. **When NOT to choose:** Don't choose this if you only need basic banking read access (official Mercury MCP suffices) or if you require functionality not yet implemented (PDF downloads, attachment uploads). ### [aws-finops-mcp-server](https://mcpcatalogs.com/en/server/ravikiranvm-aws-finops-mcp-server) by ravikiranvm · score 45 · ★176 AWS FinOps MCP server for analyzing cloud costs and detecting waste through natural language queries. **When to choose:** Choose this MCP server if you need to analyze AWS costs and detect waste across multiple profiles and regions through natural language queries, while keeping credentials secure locally. **When NOT to choose:** Don't choose this if you need support for cloud providers other than AWS, or if you're concerned about the per-query cost of the Cost Explorer API. ### [mcp-crypto-price](https://mcpcatalogs.com/en/server/truss44-mcp-crypto-price) by truss44 · score 45 · ★39 MCP server providing real-time cryptocurrency price analysis and market data via CoinCap API. **When to choose:** Choose this server when you need real-time cryptocurrency market data within MCP-enabled AI assistants for trading or investment research. **When NOT to choose:** Don't choose this if you need comprehensive trading features beyond market data, or if you prefer self-hosted solutions without external API dependencies. ### [finance-trading-ai-agents-mcp](https://mcpcatalogs.com/en/server/aitrados-finance-trading-ai-agents-mcp) by aitrados · score 45 · ★63 Professional MCP server for financial analysis and trading with real-time data, AI integration, and broker connectivity. **When to choose:** Choose this when building financial analysis systems that require real-time market data, technical indicators, or algorithmic trading capabilities integrated with LLMs. **When NOT to choose:** Avoid if you need non-financial data tools, as this server is specialized exclusively for financial operations and market data. ### [QMT-MCP](https://mcpcatalogs.com/en/server/guangxiangdebizi-qmt-mcp) by guangxiangdebizi · score 45 · ★181 QMT-MCP is a modular quantitative trading system integrating FastMCP with XTQuant for intelligent strategy generation and real-time trading execution. **When to choose:** Chinese traders using QMT/XTQuant platform who want AI-powered strategy generation and automated trading with risk controls. **When NOT to choose:** Traders using non-Chinese trading platforms, those requiring Windows compatibility, or those looking for open-source quant tools without vendor dependencies. ### [openecon-data](https://mcpcatalogs.com/en/server/hanlulong-openecon-data) by hanlulong · score 45 · ★41 MCP server providing 330K+ economic indicators from FRED, World Bank, IMF & more for AI agents. **When to choose:** Choose OpenEcon when your AI agent needs reliable economic data without hallucination, especially for macroeconomic analysis across multiple countries or when you want natural language queries instead of manual API codes. **When NOT to choose:** Don't choose OpenEcon if you need write access to economic databases, as this is a read-only service; if you're looking for non-economic time series data; or if you require offline access to the data. ### [binance-mcp-server](https://mcpcatalogs.com/en/server/analyticace-binance-mcp-server) by AnalyticAce · score 45 · ★44 A Python MCP server providing AI agents with Binance exchange tools for trading, market data, and account management. **When to choose:** Choose this MCP server when building AI agents that need to interact with Binance's cryptocurrency exchange for trading, market data, or portfolio management. **When NOT to choose:** Avoid this if you need integration with exchanges other than Binance, as it's specifically designed for Binance's API only. ### [cc-equity-research](https://mcpcatalogs.com/en/server/prof-little-bear-cc-equity-research) by prof-little-bear · score 45 · ★39 Turns Claude Code into an equity-research agent with 24 analysis skills and a unified data MCP. **When to choose:** Choose this when you need a comprehensive equity research tool that replaces expensive institutional data sources and adapts to your financial expertise level. **When NOT to choose:** Don't choose this if you need real-time data feeds beyond what drillr provides or require extensive non-US market coverage. ### [actual-mcp-server](https://mcpcatalogs.com/en/server/agigante80-actual-mcp-server) by agigante80 · score 44 · ★15 MCP server connecting Claude Desktop/LibreChat/LobeChat to Actual Budget with 63 tools for natural-language financial management. **When to choose:** Choose this server if you need comprehensive Actual Budget integration with multiple AI clients and require production-ready features like JWT authentication and concurrent connections. **When NOT to choose:** Not recommended for simple single-user local use cases with Claude Desktop only, as other simpler implementations may suffice. ### [MCP-India-Stack](https://mcpcatalogs.com/en/server/rehan1020-mcp-india-stack) by rehan1020 · score 44 · ★19 MCP server for Indian APIs with offline-first lookup tools, zero authentication, and tax calculators for AI agents. **When to choose:** Choose this server when working with Indian financial data, tax calculations, or identity validation - especially when offline access, zero authentication, and high performance for bulk operations are requirements. **When NOT to choose:** Don't choose this if you need real-time verification of registration status with government authorities, as the server only validates structural formatting and checksums. ### [quicken-mac-mcp](https://mcpcatalogs.com/en/server/dweekly-quicken-mac-mcp) by dweekly · score 44 · ★18 MCP server providing read-only access to Quicken For Mac financial data via Claude. **When to choose:** Choose this if you use Quicken For Mac and want AI-powered insights into your financial data without leaving the Claude environment. **When NOT to choose:** Don't choose this if you need write access to your financial data or if you use a different personal finance application like Mint or YNAB. ### [hive-sdk](https://mcpcatalogs.com/en/server/hive-intel-hive-sdk) by hive-intel · score 44 · ★17 Institutional-grade crypto market infrastructure MCP server with 376+ tools from 9 providers. **When to choose:** Choose Hive when you need comprehensive crypto data from multiple providers in a single normalized interface, especially for AI agents that need to answer questions across market data, DeFi, wallets, and security in one conversation. **When NOT to choose:** Avoid if you're building for environments that can't make external API calls, or if you need deep specialized data from a single provider that Hive doesn't offer normalized access to. ### [korean-dart-mcp](https://mcpcatalogs.com/en/server/chrisryugj-korean-dart-mcp) by chrisryugj · score 44 · ★36 Node.js MCP server for Korean financial disclosure APIs with 83 endpoints compressed to 15 tools. **When to choose:** Choose this MCP server if you're working with Korean financial data and need specialized tools for insider trading analysis, accounting risk assessment, or Buffett-style quality metrics directly within your AI workflow. **When NOT to choose:** Don't choose this if you need access to DART endpoints not covered by the 15 compressed tools, or if you're working with financial markets outside of Korea where the APIs are irrelevant. ### [UCAI](https://mcpcatalogs.com/en/server/nirholas-ucai) by nirholas · score 44 · ★33 UCAI is a tool that converts smart contract ABIs to MCP servers, enabling AI agents to interact with blockchain protocols. **When to choose:** When building AI applications that need to interact with blockchain contracts and want to avoid the tedious work of manually creating MCP servers. **When NOT to choose:** If you need non-EVM blockchain support, or if you require advanced features not covered by the standard ABI-to-MCP conversion. ### [universal-crypto-mcp](https://mcpcatalogs.com/en/server/nirholas-universal-crypto-mcp) by nirholas · score 44 · ★37 A universal MCP server for AI agents to interact with 20+ blockchain networks via natural language. **When to choose:** Choose this server when you need a comprehensive crypto/DeFi toolkit for AI agents that supports multiple blockchains and integrates dozens of third-party protocols. **When NOT to choose:** Avoid if you only need support for a single blockchain network or prefer minimal dependencies, as this server's extensive feature set adds complexity. ### [varrd](https://mcpcatalogs.com/en/server/augiemazza-varrd) by augiemazza · score 43 · ★17 AI-powered trading research platform with MCP server providing 9 tools for market edge validation. **When to choose:** Choose VARRD if you're a trader or quantitative researcher who needs statistically validated market analysis with transparent methodology and protection against common backtesting pitfalls. **When NOT to choose:** Avoid VARRD if you need unlimited historical data access, lower-tier pricing for casual trading research, or open-source models without vendor lock-in. ### [companies-house-mcp](https://mcpcatalogs.com/en/server/stefanoamorelli-companies-house-mcp) by stefanoamorelli · score 43 · ★18 UK Companies House API MCP server with 45+ tools for company information, search, officers, filing history, PSC, and charges. **When to choose:** Choose this MCP server when you need comprehensive UK company data through an MCP client, especially for due diligence, financial analysis, or compliance monitoring in the UK market. **When NOT to choose:** Don't choose this if you need data from outside the UK, or if you require commercial usage without contacting the author separately for licensing. ### [bitcoin-mcp](https://mcpcatalogs.com/en/server/abdelstark-bitcoin-mcp) by AbdelStark · score 43 · ★74 An MCP server enabling AI models to interact with Bitcoin and Lightning Network through various tools. **When to choose:** Choose this MCP server when you need AI agents to interact with Bitcoin and Lightning Network functionality without running a full node. **When NOT to choose:** Don't choose this if you need enterprise-level blockchain integration, advanced trading features, or support for cryptocurrencies other than Bitcoin. ### [trading-mcp](https://mcpcatalogs.com/en/server/netanelavr-trading-mcp) by netanelavr · score 43 · ★71 Trading MCP server providing stock screening, fundamental analysis, insider trading, social sentiment, and news analysis. **When to choose:** Choose this server if you need comprehensive stock analysis with multiple data sources and AI-powered sentiment analysis for investment research. **When NOT to choose:** Avoid this if you need real-time trading capabilities or if you're unwilling to manage multiple API keys for OpenAI and Reddit. ### [mcp](https://mcpcatalogs.com/en/server/twelvedata-mcp) by twelvedata · score 43 · ★66 Twelve Data MCP Server provides real-time financial market data through WebSocket, integrating seamlessly with trading applications. **When to choose:** Choose this MCP server if you need real-time financial market data integration with an AI system, especially when working with stocks, forex pairs, or cryptocurrencies. **When NOT to choose:** Avoid if you need financial data from sources other than Twelve Data, as this server creates vendor lock-in to their API and subscription services. ### [goldrush-mcp-server](https://mcpcatalogs.com/en/server/covalenthq-goldrush-mcp-server) by covalenthq · score 42 · ★14 MCP server exposing Covalent's GoldRush blockchain APIs as tools and resources for LLM clients. **When to choose:** Choose this MCP server when you need blockchain data integration for AI applications and are already using or planning to use Covalent's GoldRush API. **When NOT to choose:** Avoid this if you need blockchain data from a different provider or want to minimize dependencies on external blockchain data services. ### [aws-cost-explorer-mcp-server](https://mcpcatalogs.com/en/server/aarora79-aws-cost-explorer-mcp-server) by aarora79 · score 41 · ★127 An MCP server that provides AWS Cost Explorer and Bedrock usage data analysis through Claude Desktop or LangGraph agents. **When to choose:** Choose this MCP server if you need to analyze AWS cloud spending through natural language queries and are already using Claude Desktop. **When NOT to choose:** Avoid this if you need cost analysis for non-AWS cloud providers or require real-time cost monitoring capabilities beyond what AWS Cost Explorer provides. ### [IB_MCP](https://mcpcatalogs.com/en/server/rcontesti-ib-mcp) by rcontesti · score 41 · ★124 An MCP server that provides Interactive Brokers Web API access through Model Context Protocol for trading operations. **When to choose:** Choose IB_MCP when you need cloud-compatible IB integration without running TWS desktop software, and when you prefer standard HTTPS over TCP/IP protocols. **When NOT to choose:** Don't choose this for high-frequency trading or when you need access to all advanced TWS features that aren't yet available in the Web API. ### [TAM-MCP-Server](https://mcpcatalogs.com/en/server/gvaibhav-tam-mcp-server) by gvaibhav · score 41 · ★29 A comprehensive MCP server for market sizing analysis with 28 tools, 15 business prompts, and integration with 8 economic data sources. **When to choose:** Choose this server for professional market analysis and business intelligence work, especially when you need TAM/SAM calculations, industry research, or integration with economic data sources. **When NOT to choose:** Not ideal if you need real-time market data with minimal latency, as the server relies on external APIs with potential caching delays, or if you require highly specialized industry data not covered by the integrated sources. ### [mcp-server](https://mcpcatalogs.com/en/server/finmap-org-mcp-server) by finmap-org · score 41 · ★10 A financial data MCP server providing stock exchange data from multiple countries with visualization capabilities. **When to choose:** Choose this MCP server when you need comprehensive historical stock market data from multiple international exchanges and want seamless integration with AI assistants. **When NOT to choose:** Avoid this if you need real-time trading data, extended market coverage beyond the supported exchanges, or have compliance concerns with accessing international financial data. ### [mcp-server](https://mcpcatalogs.com/en/server/paperinvest-mcp-server) by paperinvest · score 40 · ★21 Official MCP server for Paper's trading platform enabling AI assistants to interact with trading APIs. **When to choose:** Choose this server if you're using Paper's trading platform and want to integrate AI assistants with its API for simulated trading activities. **When NOT to choose:** Don't choose this if you need live trading capabilities, as it's designed specifically for paper trading only. ### [polymarket-mcp](https://mcpcatalogs.com/en/server/pab1it0-polymarket-mcp) by pab1it0 · score 40 · ★6 MCP server providing access to Polymarket's prediction markets data through standardized interfaces. **When to choose:** Choose this MCP server if you need AI assistants to access Polymarket's prediction market data through standardized interfaces without implementing custom API integrations. **When NOT to choose:** Don't choose this if you need access to other prediction markets beyond Polymarket or require more advanced trading functionality than market data access. ### [qmt-mcp-server](https://mcpcatalogs.com/en/server/jm12138-qmt-mcp-server) by jm12138 · score 40 · ★18 An MCP server for QMT platform that provides stock market data download and query functionality. **When to choose:** Choose this MCP server if you need access to Chinese stock market data through QMT platform for financial analysis and research. **When NOT to choose:** Avoid if you need cross-market data (US, Europe, etc.) as this server is limited to Chinese stocks and Windows platform only. ### [solscan-mcp](https://mcpcatalogs.com/en/server/wowinter13-solscan-mcp) by wowinter13 · score 39 · ★40 A Rust MCP server for querying Solana blockchain data through Solscan API using natural language. **When to choose:** Choose this MCP server when you need to query Solana blockchain data through natural language and want reliable error handling for investigative workflows. **When NOT to choose:** Don't choose this if you need to work with other blockchain networks besides Solana, as it's specifically designed for Solana's ecosystem only. ### [monobank-mcp](https://mcpcatalogs.com/en/server/aler1x-monobank-mcp) by Aler1x · score 38 · ★2 Node.js/TypeScript MCP server for Monobank that exposes personal accounts, statements, and public exchange rates as tools. **When to choose:** Choose this if you're working in a Node.js/TypeScript ecosystem and need to access Monobank account data or public currency exchange rates through MCP. **When NOT to choose:** Don't choose this if you need a Python implementation or if you're already using the original Python version with no TypeScript requirements. ### [wealthsimple-mcp](https://mcpcatalogs.com/en/server/tatianathevisionary-wealthsimple-mcp) by tatianathevisionary · score 37 · ★1 TypeScript MCP server connecting AI agents to Wealthsimple's help center via public Zendesk API with 7 typed tools. **When to choose:** Choose this server when you need to provide AI-powered support for Wealthsimple products with access to their official help center documentation. **When NOT to choose:** Don't use this if you need account-specific data or real-time transaction information, as it only provides access to the public help center documentation. ### [polymarket-mcp](https://mcpcatalogs.com/en/server/aryankeluskar-polymarket-mcp) by aryankeluskar · score 37 · ★6 A comprehensive MCP server for accessing Polymarket's prediction markets through Claude AI with market discovery, analysis tools, and real-time data. **When to choose:** Choose this server when you need to access prediction market data through Claude for market analysis, trading insights, or educational content creation. **When NOT to choose:** Don't choose this server if you need authenticated access to private market data or if you're looking for a solution with built-in trading capabilities rather than read-only market analysis. ### [your-money-left-the-chat](https://mcpcatalogs.com/en/server/rayato159-your-money-left-the-chat) by Rayato159 · score 37 · ★23 A Rust + MCP powered financial tracker that logs expenses and analyzes habits through natural conversation with local SQLite storage. **When to choose:** Choose this for personal finance tracking when you value privacy and want to interact through natural conversation with AI assistants. **When NOT to choose:** Not suitable for business finance management or if you need multi-user support, as it's designed for personal use with single-user SQLite storage. ### [beanquery-mcp](https://mcpcatalogs.com/en/server/vanto-beanquery-mcp) by vanto · score 36 · ★46 Experimental MCP server for AI assistants to query Beancount financial ledgers using BQL. **When to choose:** Choose this server if you use Beancount for financial tracking and want AI assistants to query your ledger data without manual extraction. **When NOT to choose:** Avoid this server if your financial data contains highly sensitive information and you cannot use self-hosted LLMs, as data is transmitted to third-party services. ### [polymarket-mcp-server](https://mcpcatalogs.com/en/server/justkokosixnine-polymarket-mcp-server) by Justkokosixnine · score 36 · ★3 An MCP server providing AI-powered tools for Polymarket prediction market trading with real-time monitoring. **When to choose:** Choose this MCP server if you're actively trading on Polymarket and want AI assistance with market analysis and real-time monitoring. **When NOT to choose:** Don't choose this if you're not using Polymarket for prediction markets or if you prefer a self-hosted solution without pre-built binaries. ### [educhain-ai-agent-kit](https://mcpcatalogs.com/en/server/sailfish-finance-educhain-ai-agent-kit) by SailFish-Finance · score 36 · ★6 MCP server for EDUCHAIN blockchain operations including token queries, wallet management, swaps, and arbitrage detection. **When to choose:** Choose this server when working with EDUCHAIN ecosystem, SailFish DEX, or when you need DeFi tools for token swaps and arbitrage opportunities. **When NOT to choose:** Avoid if you're not working with EDUCHAIN or SailFish DEX specifically, as this server is tightly integrated with this particular ecosystem. ### [korean-crypto-mcp](https://mcpcatalogs.com/en/server/koreafintech-korean-crypto-mcp) by koreafintech · score 36 · ★0 MCP server providing real-time Korean cryptocurrency market data, including kimchi premium tracking and exchange comparisons. **When to choose:** Choose this MCP server if you need real-time Korean cryptocurrency market data, especially for tracking kimchi premium and arbitrage opportunities between Korean exchanges. **When NOT to choose:** Don't choose this if you need data from exchanges outside South Korea or if you require more comprehensive market coverage beyond the KRW markets supported by Upbit. ### [botindex-mcp-server](https://mcpcatalogs.com/en/server/cyberweasel777-botindex-mcp-server) by Cyberweasel777 · score 36 · ★0 BotIndex MCP Server provides AI agents with sports odds, crypto data, and analytics via 17 premium tools. **When to choose:** Choose this when you need real-time sports betting intelligence, cryptocurrency analytics, or token launch data for your AI agents, especially if you're already familiar with or interested in implementing x402 payment protocols. **When NOT to choose:** Avoid if you need free, unlimited access to premium data or prefer traditional API key authentication over wallet-based payment models. ### [zerodha-mcp](https://mcpcatalogs.com/en/server/mtwn105-zerodha-mcp) by mtwn105 · score 35 · ★30 A Python-based MCP server for Zerodha trading with interactive chat interface and multi-agent support. **When to choose:** Choose this MCP server if you need AI-powered interaction with a Zerodha trading account and want multiple client options including web interfaces. **When NOT to choose:** Don't choose this if you're using a different brokerage platform, as it's specifically designed for Zerodha and won't work with other trading services. ### [mcp-yfinance](https://mcpcatalogs.com/en/server/9nate-drake-mcp-yfinance) by 9nate-drake · score 34 · ★19 MCP server providing Yahoo Finance data through yfinance API to Claude Desktop. **When to choose:** Choose this when you need financial data access directly in Claude Desktop without switching between applications. **When NOT to choose:** Don't choose this if you need highly specialized financial analysis tools or data from sources beyond Yahoo Finance. ### [mcp-free-usdc-transfer](https://mcpcatalogs.com/en/server/magnetai-mcp-free-usdc-transfer) by magnetai · score 34 · ★20 MCP server for free USDC transfers on Base blockchain using Coinbase CDP wallets. **When to choose:** Choose this when you need an AI system to perform free USDC transfers on Base chain without worrying about transaction fees. **When NOT to choose:** Don't choose this if you need transfers on other blockchains, or if you're concerned about vendor lock-in with Coinbase CDP. ### [Finance-personal-assistant](https://mcpcatalogs.com/en/server/beprith-finance-personal-assistant) by beprith · score 32 · ★2 A local MCP server providing mock financial data tools for testing applications without connecting to real systems. **When to choose:** Developers needing to test financial applications in a secure, isolated environment without accessing real financial data or APIs. **When NOT to choose:** Production systems requiring real-time financial data or complex transaction processing capabilities. ### [solscan-mcp](https://mcpcatalogs.com/en/server/valennmg-solscan-mcp) by Valennmg · score 32 · ★2 An MCP server for querying Solana blockchain data using natural language through the Solscan API. **When to choose:** Choose this if you need a basic Rust implementation for simple Solana transaction queries and are willing to contribute to its development. **When NOT to choose:** Don't choose this for production use as it lacks proper MCP integration documentation and appears to be in early development stages. ### [Airbnb-StockScreener-Agent-with-MCP](https://mcpcatalogs.com/en/server/ambidextrous9-airbnb-stockscreener-agent-with-mcp) by ambideXtrous9 · score 27 · ★0 A local stock analysis MCP server using LangChain, Ollama Qwen3, and BeautifulSoup for financial data. **When to choose:** Choose this for local stock analysis without requiring external API calls, particularly when working with Indian stocks. **When NOT to choose:** Avoid if you need real-time stock data or analysis of international markets beyond what Screener.in covers. --- ## games (1) ### [bgg-mcp](https://mcpcatalogs.com/en/server/kkjdaniel-bgg-mcp) by kkjdaniel · score 45 · ★40 BGG MCP provides BoardGameGeek API access through MCP with tools for searching, collections, pricing, and recommendations. **When to choose:** Choose BGG MCP if you're a board game enthusiast who wants to integrate BGG data into AI workflows for collection management, trading, or research. **When NOT to choose:** Avoid if you need extensive write capabilities to BGG, as this server is primarily read-only, or if you prefer a language other than Go for your MCP servers. --- ## github (9) ### [github-mcp-server](https://mcpcatalogs.com/en/server/github-github-mcp-server) by github · score 64 · ★29,881 GitHub's official MCP Server connects AI tools to GitHub's platform for repository management, issue automation, and code analysis. **When to choose:** Choose this when you need direct GitHub integration for AI tools, especially if you're already using GitHub Copilot or want to leverage GitHub's official implementation for better compatibility. **When NOT to choose:** Avoid if you need self-hosting without internet access, require offline functionality, or need to integrate with other version control systems beyond GitHub. ### [gistpad-mcp](https://mcpcatalogs.com/en/server/lostintangent-gistpad-mcp) by lostintangent · score 47 · ★192 MCP server for managing knowledge, notes and prompts via GitHub Gists with extensive tooling for gist operations. **When to choose:** Choose GistPad MCP if you already use GitHub Gists for knowledge management and want to integrate them directly into your AI workflow. **When NOT to choose:** Don't choose this if you need more advanced GitHub features beyond gists, or if you prefer not to store your knowledge in GitHub's ecosystem. ### [mcp-github-project-manager](https://mcpcatalogs.com/en/server/kunwarvivek-mcp-github-project-manager) by kunwarVivek · score 46 · ★88 An MCP server for GitHub project management with AI-powered task generation and requirements traceability. **When to choose:** Choose this MCP server if you need AI-powered GitHub project management with complete requirements traceability and are already using GitHub Projects v2. **When NOT to choose:** Avoid if you need a lightweight solution without AI dependencies, or if you're not using GitHub's Project management features (v2). ### [github-brain](https://mcpcatalogs.com/en/server/wham-github-brain) by wham · score 46 · ★75 GitHub Brain MCP server summarizes GitHub discussions, issues, and PRs with local database for fast responses. **When to choose:** Choose GitHub Brain when you need to analyze large amounts of GitHub data beyond the standard API limits, require faster response times, or prefer to keep data local for security reasons. **When NOT to choose:** Avoid this if you need real-time data from GitHub or require write access to repositories, as this is a read-only solution that requires periodic database updates. ### [mcp-git-ingest](https://mcpcatalogs.com/en/server/adhikasp-mcp-git-ingest) by adhikasp · score 44 · ★306 A Model Context Protocol server that reads GitHub repository structures and important files. **When to choose:** When you need to quickly analyze GitHub repository structures or read specific files as part of an AI workflow, especially for public repositories. **When NOT to choose:** When working with private repositories requiring authentication, or when analyzing repositories with large binary files that might exceed memory limits. ### [Gitingest-MCP](https://mcpcatalogs.com/en/server/puravparab-gitingest-mcp) by puravparab · score 42 · ★137 MCP server for extracting GitHub repository summaries, structure, and content. **When to choose:** Choose this when you need rapid analysis of GitHub repositories during AI-assisted development, especially when working with unfamiliar codebases. **When NOT to choose:** Avoid if you need to analyze private repositories not accessible via public GitHub APIs, or if you require more granular repository analysis tools. ### [github-chat-mcp](https://mcpcatalogs.com/en/server/asyncfuncai-github-chat-mcp) by AsyncFuncAI · score 41 · ★86 An MCP server for analyzing GitHub repositories using the GitHub Chat API. **When to choose:** Choose this MCP server when you need to analyze and query GitHub repositories through conversational AI interfaces, especially if you're already using GitHub Chat API or need repository indexing capabilities. **When NOT to choose:** Avoid this server if you need to work with private repositories without a GitHub Chat API key, or if you require functionality beyond repository analysis and querying. ### [github-mcp-server](https://mcpcatalogs.com/en/server/parassolanki-github-mcp-server) by ParasSolanki · score 32 · ★4 MCP server for GitHub API integration, providing extensive repository, issue, PR, and code search capabilities. **When to choose:** Choose this MCP server if you need GitHub repository management and search functionality integrated into your LLM workflows, particularly for reading repository data, issues, and PRs. **When NOT to choose:** Don't choose this server if you need GitHub write operations (it's read-only), prefer more comprehensive GitHub automation, or require authentication methods beyond personal access tokens. ### [GitHub-URL-MCP](https://mcpcatalogs.com/en/server/kongyo2-github-url-mcp) by kongyo2 · score 31 · ★1 An MCP server that validates and converts GitHub URLs between repository info and properly formatted URLs. **When to choose:** Choose this when you need reliable GitHub URL validation and parsing without authentication requirements, particularly for AI systems that struggle with GitHub URL handling. **When NOT to choose:** Not ideal if you need deep repository analysis beyond URL validation, or if you require authentication for private repository access. --- ## health (2) ### [whoop-mcp](https://mcpcatalogs.com/en/server/shashankswe2020-ux-whoop-mcp) by shashankswe2020-ux · score 49 · ★70 MCP server connecting AI assistants to WHOOP health and fitness data through OAuth2 authentication with 6 data tools. **When to choose:** Choose this MCP server if you're a WHOOP user who wants to integrate your health data with AI assistants for natural language analysis of your fitness metrics. **When NOT to choose:** Don't choose this if you don't have a WHOOP membership or need access to other health data sources not supported by WHOOP. ### [garmin-connect-mcp](https://mcpcatalogs.com/en/server/eddmann-garmin-connect-mcp) by eddmann · score 41 · ★36 A fully-featured MCP server providing comprehensive tools for Garmin Connect data access. **When to choose:** Choose this server if you're a fitness enthusiast who wants to analyze your Garmin data through LLMs without leaving your chat interface. **When NOT to choose:** Avoid if you need to access data from other fitness platforms, as this only works with Garmin Connect. --- ## healthcare (3) ### [clinicaltrialsgov-mcp-server](https://mcpcatalogs.com/en/server/cyanheads-clinicaltrialsgov-mcp-server) by cyanheads · score 48 · ★71 MCP server providing 7 tools for searching, retrieving, and matching clinical trials from ClinicalTrials.gov API. **When to choose:** Choose this server when you need to search clinical trials, match patients to studies, or analyze trial data in AI workflows, especially if you want immediate access without hosting your own instance. **When NOT to choose:** Don't choose this if you need to modify or create clinical trial data (this is read-only only), or if you require highly sensitive healthcare data handling with specific compliance requirements. ### [healthcare-mcp-public](https://mcpcatalogs.com/en/server/cicatriiz-healthcare-mcp-public) by Cicatriiz · score 44 · ★115 Healthcare MCP Server provides AI assistants with access to authoritative medical data including FDA drugs, PubMed, clinical trials, and medical calculators. **When to choose:** Choose this server if you need to build medical AI assistants that require authoritative, up-to-date healthcare information from official sources like FDA, PubMed, and clinical trials databases. **When NOT to choose:** Not ideal for healthcare applications requiring real-time patient data or sensitive medical record access, as this server only provides access to public medical databases and reference information. ### [fhir-mcp-server](https://mcpcatalogs.com/en/server/the-momentum-fhir-mcp-server) by the-momentum · score 43 · ★80 A production-ready MCP server for FHIR healthcare data integration with Claude Desktop. **When to choose:** Choose this server when working with healthcare data and need natural language access to FHIR-compliant systems with secure authentication. **When NOT to choose:** Avoid if you don't have access to a FHIR server or need broader healthcare data management beyond what FHIR provides. --- ## home-automation (2) ### [hass-mcp](https://mcpcatalogs.com/en/server/voska-hass-mcp) by voska · score 48 · ★296 Hass-MCP enables AI assistants to interact with Home Assistant through device control, entity queries, and automation management. **When to choose:** Choose Hass-MCP if you want AI-powered control of your Home Assistant smart home devices through natural language interactions. **When NOT to choose:** Don't choose Hass-MCP if you use a smart home platform other than Home Assistant or need advanced automation not supported by the available tools. ### [mcp-server-home-assistant](https://mcpcatalogs.com/en/server/allenporter-mcp-server-home-assistant) by allenporter · score 37 · ★67 MCP server for Home Assistant that enables AI models to interact with home automation systems. **When to choose:** Choose this if you use Claude Desktop and want to control or query your Home Assistant smart home devices through natural language. **When NOT to choose:** Don't choose this if you're not using Home Assistant, need more comprehensive home automation features, or require read/write access beyond what Home Assistant provides. --- ## iot (1) ### [tinymcp](https://mcpcatalogs.com/en/server/golioth-tinymcp) by golioth · score 43 · ★150 MCP server enabling LLMs to control embedded devices via Golioth API and RPCs. **When to choose:** Choose tinymcp when you need AI systems to interact with physical embedded devices and you're already using or willing to adopt the Golioth platform. **When NOT to choose:** Avoid tinymcp if you don't want vendor lock-in to Golioth, need direct device control without an intermediary platform, or require production stability as it's still experimental. --- ## knowledge-graph (40) ### [memory](https://mcpcatalogs.com/en/server/modelcontextprotocol-servers-memory) by modelcontextprotocol · score 77 · ★85,748 · official An MCP server implementing persistent memory using a local knowledge graph for AI models to remember user information across chats. **When to choose:** Choose this server if you need persistent memory for an AI assistant to remember user information across chat sessions, especially when working with TypeScript-based environments or need a self-hosted solution without external dependencies. **When NOT to choose:** Avoid this for high-memory systems requiring complex reasoning capabilities beyond simple facts and relationships, or if you need integration with existing databases rather than file-based storage. ### [biomcp](https://mcpcatalogs.com/en/server/genomoncology-biomcp) by genomoncology · score 53 · ★507 BioMCP is a biomedical-focused MCP server that provides unified access to multiple biomedical data sources through a single command interface. **When to choose:** Choose BioMCP when you need unified access to multiple biomedical databases and literature sources for research, clinical queries, or bioinformatics analysis without learning multiple APIs. **When NOT to choose:** Don't choose BioMCP if you need write access to databases (it's read-only), or if you work primarily with non-biomedical data domains outside its specialized scope. ### [arkon](https://mcpcatalogs.com/en/server/nduckmink-arkon) by nduckmink · score 52 · ★729 Enterprise AI knowledge hub serving as MCP server for secure organizational RAG contexts and AI skills. **When to choose:** Choose Arkon for enterprise knowledge management with strict isolation requirements across departments and projects, especially when you need a comprehensive MCP solution with human review workflows. **When NOT to choose:** Don't choose Arkon if you need a permissive license that allows SaaS hosting or want a simpler, lightweight solution for small teams without complex RBAC requirements. ### [graphlit-mcp-server](https://mcpcatalogs.com/en/server/graphlit-graphlit-mcp-server) by graphlit · score 51 · ★375 MCP server connecting Graphlit platform with rich tools for knowledge management and RAG across various data sources. **When to choose:** Teams already using Graphlit who want to integrate it with MCP-compatible clients like Cursor or Windsurf for enhanced knowledge retrieval and RAG capabilities. **When NOT to choose:** Organizations concerned about vendor lock-in to Graphlit's ecosystem, or those who need a self-hosted solution without external dependencies. ### [swarmvault](https://mcpcatalogs.com/en/server/swarmclawai-swarmvault) by swarmclawai · score 51 · ★454 SwarmVault is an MCP server that turns documents into a local knowledge graph with RAG capabilities. **When to choose:** Choose SwarmVault when you need a local, persistent knowledge graph for AI agents that works offline and requires no API keys. **When NOT to choose:** Avoid SwarmVault if you need cloud-based, real-time collaborative knowledge management with multiple users editing simultaneously. ### [mcp-logseq](https://mcpcatalogs.com/en/server/ergut-mcp-logseq) by ergut · score 51 · ★267 An MCP server connecting Claude to LogSeq knowledge bases with read/write capabilities and optional vector search. **When to choose:** Choose this if you're already using LogSeq and want to integrate AI directly into your knowledge workflow without data export or context switching. **When NOT to choose:** Don't choose this if you need cross-platform support beyond LogSeq, or if you're uncomfortable with running a local HTTP API server with full access to your notes. ### [turbovault](https://mcpcatalogs.com/en/server/epistates-turbovault) by Epistates · score 50 · ★126 Rust SDK and MCP server that transforms Obsidian vaults into intelligent knowledge systems with 47 specialized tools. **When to choose:** Choose TurboVault if you need advanced knowledge management features for Obsidian vaults with high-performance tool calling and graph analysis capabilities. **When NOT to choose:** Avoid if you're looking for a lightweight solution without Rust dependencies or if you need write operations beyond the provided atomic batch capabilities. ### [anytype-mcp](https://mcpcatalogs.com/en/server/anyproto-anytype-mcp) by anyproto · score 50 · ★420 MCP server for AI assistants to interact with Anytype's encrypted wiki through natural language commands. **When to choose:** Choose this MCP server if you use Anytype as your primary knowledge management system and want to interact with it through AI assistants without switching contexts. **When NOT to choose:** Don't choose this if you don't use Anytype, as it requires an Anytype installation and API key for functionality. ### [MegaMemory](https://mcpcatalogs.com/en/server/0xk3vin-megamemory) by 0xK3vin · score 50 · ★168 MegaMemory is an MCP server that creates persistent project knowledge graphs with semantic search for coding agents. **When to choose:** Choose MegaMemory when your coding agents need persistent memory across sessions without complex AST parsing, especially for projects where architectural understanding evolves over time. **When NOT to choose:** Avoid MegaMemory if you need real-time code symbol analysis or have projects with strict zero-disk-usage requirements, as it stores all knowledge locally in SQLite. ### [Corbell](https://mcpcatalogs.com/en/server/corbell-ai-corbell) by Corbell-AI · score 50 · ★208 MCP server providing architecture graph intelligence, code search, and spec generation tools for backend teams. **When to choose:** Teams with complex multi-repo architectures needing AI-powered documentation and specification generation that respects established patterns. **When NOT to choose:** Teams working on small, single-repo projects without significant architectural complexity or needing real-time collaboration features. ### [memex](https://mcpcatalogs.com/en/server/iamtouchskyer-memex) by iamtouchskyer · score 49 · ★201 Zettelkasten-based persistent memory system for AI coding agents with MCP server integration. **When to choose:** Choose Memex if you use multiple AI coding tools and want persistent memory that works across all of them with zero configuration. **When NOT to choose:** Don't choose Memex if you need advanced search capabilities beyond basic text search or require enterprise-grade security features. ### [memento-mcp](https://mcpcatalogs.com/en/server/gannonh-memento-mcp) by gannonh · score 49 · ★418 Knowledge graph memory system with semantic retrieval and temporal awareness for LLMs. **When to choose:** Choose Memento MCP when you need persistent knowledge graph memory with semantic retrieval capabilities for LLM applications, especially when you're already using or willing to set up Neo4j. **When NOT to choose:** Don't choose Memento MCP if you need a lightweight solution without additional database dependencies or if you require graph operations without vector search capabilities. ### [MediaWiki-MCP-Server](https://mcpcatalogs.com/en/server/professionalwiki-mediawiki-mcp-server) by ProfessionalWiki · score 49 · ★93 MCP server connecting AI with MediaWiki wikis through tools and resources for reading and writing content. **When to choose:** Choose this server when you need to integrate AI with MediaWiki instances for content management, especially when working with specialized MediaWiki extensions like Cargo or Semantic MediaWiki. **When NOT to choose:** Avoid this if you need to interact with non-MediaWiki wikis or if your use case requires extensive write operations without proper authentication setup. ### [open-ontologies](https://mcpcatalogs.com/en/server/fabio-rovai-open-ontologies) by fabio-rovai · score 49 · ★110 AI-native ontology engine Rust MCP server with 43 tools for building, validating, and querying RDF/OWL ontologies using Oxigraph triple store. **When to choose:** Choose Open Ontologies when working with complex ontologies and knowledge graphs that require AI-assisted building, validation, and reasoning capabilities without the overhead of Java-based tools like Protégé. **When NOT to choose:** Avoid if you need collaborative ontology editing features, as the current version appears to be designed for single-user environments without apparent multi-user support. ### [atlas-mcp-server](https://mcpcatalogs.com/en/server/cyanheads-atlas-mcp-server) by cyanheads · score 48 · ★475 A Neo4j-powered MCP server implementing three-tier project/task/knowledge management for LLM agents with deep research capabilities. **When to choose:** Choose ATLAS when you need a structured knowledge graph for complex project management with dependency tracking for LLM agents. **When NOT to choose:** Avoid ATLAS if you need a lightweight solution without external database dependencies or if you prefer open-source relational databases over Neo4j. ### [ols4](https://mcpcatalogs.com/en/server/ebispot-ols4) by EBISPOT · score 47 · ★89 EMBL-EBI's Ontology Lookup Service provides an MCP server for biomedical ontologies via REST API. **When to choose:** Choose OLS when you need access to standardized biomedical ontologies through MCP, especially for AI applications in bioinformatics or healthcare. **When NOT to choose:** Not suitable if you need ontology reasoning capabilities, as OLS assumes pre-reasoned ontologies and doesn't perform any OWL reasoning. ### [mcp-server-atlassian-confluence](https://mcpcatalogs.com/en/server/aashari-mcp-server-atlassian-confluence) by aashari · score 47 · ★54 MCP server connecting AI assistants to Atlassian Confluence knowledge bases with API tools for spaces, pages, and search. **When to choose:** Teams using Atlassian Confluence who want to connect their knowledge base directly to AI assistants without complex integrations. **When NOT to choose:** You need write permissions in Confluence, as the server primarily offers read-only access with limited content creation capabilities. ### [flexible-graphrag](https://mcpcatalogs.com/en/server/stevereiner-flexible-graphrag) by stevereiner · score 47 · ★127 MCP server providing tools for document ingestion, hybrid search, and AI queries in Flexible GraphRAG. **When to choose:** Choose Flexible GraphRAG MCP server when you need to connect Claude Desktop to a comprehensive document processing and knowledge graph platform with support for multiple database backends. **When NOT to choose:** Avoid if you need a lightweight solution or if you're not already using or planning to use one of the supported databases and frameworks. ### [mie](https://mcpcatalogs.com/en/server/kraklabs-mie) by kraklabs · score 46 · ★42 MIE is a persistent memory graph MCP server that enables AI agents to share and recall knowledge across sessions. **When to choose:** Choose MIE when you need persistent memory across multiple AI agents and want structured knowledge storage with relationship tracking. **When NOT to choose:** Don't choose MIE if you need a simple chat history replacement or if you require a cloud-based solution with multi-device synchronization (not yet available). ### [dreamgraph](https://mcpcatalogs.com/en/server/mmethodz-dreamgraph) by mmethodz · score 46 · ★82 DreamGraph is a graph-first cognitive layer that builds a persistent knowledge graph to reason about code architecture. **When to choose:** Choose DreamGraph if you need to maintain architectural memory across multiple repositories and want a persistent knowledge graph for reasoning about code changes and system architecture. **When NOT to choose:** Avoid DreamGraph if you need lightweight integration with existing tools, have limited resources to run additional services, or prefer a simpler solution without persistent graph storage. ### [ontosphere](https://mcpcatalogs.com/en/server/thhanke-ontosphere) by ThHanke · score 46 · ★59 Browser-based RDF knowledge graph editor with MCP server integration for AI agent workflows. **When to choose:** Choose Ontosphere when you need a browser-based RDF/ontology editor with AI integration that works entirely client-side without requiring server infrastructure. **When NOT to choose:** Avoid Ontosphere if you need to work with very large knowledge graphs that exceed browser memory limits or require server-side processing capabilities. ### [brain-in-the-fish](https://mcpcatalogs.com/en/server/fabio-rovai-brain-in-the-fish) by fabio-rovai · score 46 · ★66 An MCP server that scores documents by verifying claims against source text through OWL knowledge graphs. **When to choose:** Choose this tool when you need to validate documents for evidence-backed claims, particularly in domains like tender evaluation, academic grading, or policy review where proving scores is essential. **When NOT to choose:** Avoid using this for real-time processing of large volumes of documents due to its local-only architecture requiring Rust compilation and limited scalability for batch operations. ### [second-brain-cloudflare](https://mcpcatalogs.com/en/server/rahilp-second-brain-cloudflare) by rahilp · score 46 · ★61 Self-hosted MCP server providing personal memory layer across AI tools like Claude and ChatGPT. **When to choose:** Choose this MCP server if you want a consistent memory layer across multiple AI tools and are looking for a simple, self-hosted solution that requires minimal technical setup. **When NOT to choose:** Avoid this if you need enterprise-grade security features, require high-scale data processing, or want the ability to easily migrate data between different providers. ### [mcp-duckdb-memory-server](https://mcpcatalogs.com/en/server/izumisy-mcp-duckdb-memory-server) by IzumiSy · score 46 · ★57 A production-ready MCP memory server using DuckDB as backend for knowledge graph storage and retrieval. **When to choose:** Choose this when you need a persistent, scalable memory system for AI applications with structured knowledge graphs and efficient search capabilities. **When NOT to choose:** Don't choose this if you need a simple in-memory solution for very small datasets or if you prefer JSON-based storage without external dependencies. ### [compass](https://mcpcatalogs.com/en/server/raystack-compass) by raystack · score 46 · ★67 Compass is a knowledge graph context engine that provides MCP tools for AI agents to search and traverse organizational metadata. **When to choose:** Choose Compass when you need to map complex organizational systems and want AI agents to understand relationships and impact across your infrastructure. **When NOT to choose:** Don't choose Compass if you need real-time updates (it's focused on temporal knowledge, not live data) or if you're already heavily invested in a different graph database ecosystem. ### [mcp-open-library](https://mcpcatalogs.com/en/server/8ensmith-mcp-open-library) by 8enSmith · score 46 · ★71 An MCP server providing book and author information search via Open Library API with comprehensive tools for AI assistants. **When to choose:** Choose this MCP server when your AI assistant needs to provide book recommendations, search for author information, or access bibliographic data from the Open Library catalog. **When NOT to choose:** Don't choose this if you need access to commercial book databases like Google Books or have specialized library system requirements beyond Open Library's scope. ### [flywheel-memory](https://mcpcatalogs.com/en/server/velvetmonkey-flywheel-memory) by velvetmonkey · score 45 · ★44 Local-first MCP server that provides AI agents with a knowledge graph over Obsidian vaults. **When to choose:** Choose Flywheel when you need an AI memory layer for Obsidian or plain Markdown files, with local-first architecture, safe bounded writes, and optional semantic search. **When NOT to choose:** Don't choose Flywheel if you need cloud-based memory access, require write access to non-Markdown formats, or need multi-user collaborative features. ### [nowledge-mem](https://mcpcatalogs.com/en/server/nowledge-co-nowledge-mem) by nowledge-co · score 45 · ★181 A local-first, graph-based memory manager that integrates with AI tools via MCP protocol. **When to choose:** Choose Nowledge Mem when you need a unified memory system for AI interactions across multiple tools, especially if you value knowledge graph organization. **When NOT to choose:** Avoid if you need completely offline functionality without any server components, as this requires a local server running on port 14242. ### [eion](https://mcpcatalogs.com/en/server/eiondb-eion) by eiondb · score 45 · ★155 Eion provides a shared memory storage with knowledge graph capabilities for multi-agent systems via MCP server. **When to choose:** Choose Eion when you need shared memory with knowledge graph capabilities for multi-agent systems, especially when already using or comfortable with PostgreSQL and Neo4j as the underlying data stores. **When NOT to choose:** Avoid Eion if you need a simpler solution without graph capabilities, if you want to avoid the complexity of maintaining two separate database systems, or if you prefer a managed cloud service over self-hosted infrastructure. ### [cuba-memorys](https://mcpcatalogs.com/en/server/leandropg19-cuba-memorys) by LeandroPG19 · score 45 · ★22 A sophisticated MCP server providing AI agents with persistent memory using knowledge graphs, neuroscience-inspired algorithms, and anti-hallucination techniques. **When to choose:** Choose Cuba-Memorys when you need neuroscience-inspired persistent memory with advanced features like knowledge graphs, contradiction detection, and anti-hallucination grounding for AI agents. **When NOT to choose:** Not suitable if you need a simple memory solution without heavy dependencies, or if you require a non-CC BY-NC 4.0 licensed solution for commercial use. ### [opentargets-mcp](https://mcpcatalogs.com/en/server/nickzren-opentargets-mcp) by nickzren · score 45 · ★17 MCP server providing access to Open Targets biomedical data through 68 curated tools and 3 advanced GraphQL tools. **When to choose:** Choose this server for biomedical research when you need structured access to Open Targets' comprehensive genetic, disease, and drug data through well-curated tools. **When NOT to choose:** Don't choose this if you need write access to the database (it's read-only) or if you're working with non-biomedical domains. ### [WikidataMCP](https://mcpcatalogs.com/en/server/wmde-wikidatamcp) by wmde · score 43 · ★24 MCP server providing standardized tools to explore and query Wikidata programmatically for AI applications. **When to choose:** Choose WikidataMCP when your AI application needs to explore and query Wikidata programmatically, especially when leveraging semantic search or SPARQL queries. **When NOT to choose:** Don't choose this if you need write access to Wikidata (this is read-only) or if you require more comprehensive Wikidata integration beyond query capabilities. ### [segundo-cerebro](https://mcpcatalogs.com/en/server/orobsonn-segundo-cerebro) by orobsonn · score 43 · ★19 A personal knowledge graph MCP server for Claude that runs on Cloudflare with automatic concept atomization and cross-domain analogy connections. **When to choose:** Choose this if you work across multiple domains and want Claude to help build a knowledge graph with substantive cross-domain connections, already use Claude Code/Desktop regularly, and are on Pro, Max, or API plans. **When NOT to choose:** Don't choose this if you're on Claude Free tier, need offline access, work within a single narrow domain, or prefer traditional hierarchical note organization over lattice-based connections. ### [ContextGraph](https://mcpcatalogs.com/en/server/allenmaxi-contextgraph) by AllenMaxi · score 43 · ★21 MCP server implementing a shared memory bus with permissions, subscriptions, and optional payments for multi-agent teams. **When to choose:** Choose ContextGraph when you need shared memory for multiple coding agents with access controls and structured state preservation. **When NOT to choose:** Avoid if you only need personal memory for a single chatbot or if you require hosted agent runtime with enterprise IAM capabilities. ### [zettelkasten-mcp](https://mcpcatalogs.com/en/server/entanglr-zettelkasten-mcp) by entanglr · score 42 · ★151 Zettelkasten MCP server enables atomic note creation, linking, and exploration through MCP protocol for knowledge management. **When to choose:** Choose this server if you're implementing the Zettelkasten method for personal knowledge management and want AI-powered exploration of your notes through MCP integration. **When NOT to choose:** Don't choose this if you need real-time collaboration features, as it's designed for single-user knowledge management. ### [PDB-MCP-Server](https://mcpcatalogs.com/en/server/augmented-nature-pdb-mcp-server) by Augmented-Nature · score 42 · ★24 An MCP server providing access to Protein Data Bank structures with search, retrieval and download capabilities. **When to choose:** Choose this MCP server if you need direct access to Protein Data Bank structures for molecular biology research and have existing workflows that benefit from programmatic access to PDB data. **When NOT to choose:** Don't choose this if you need real-time visualization capabilities or require access to experimental protein structure prediction data not yet in PDB. ### [owl-mcp](https://mcpcatalogs.com/en/server/ai4curation-owl-mcp) by ai4curation · score 39 · ★15 OWL-MCP enables AI assistants to interact with OWL ontologies via standardized MCP protocol. **When to choose:** Researchers and knowledge engineers working with OWL ontologies who want AI assistance for creating, modifying, and maintaining ontologies. **When NOT to choose:** If you need advanced ontology visualization features or support for other knowledge representation formats besides OWL. ### [macrostrat-mcp](https://mcpcatalogs.com/en/server/blake365-macrostrat-mcp) by blake365 · score 38 · ★7 MCP server for accessing comprehensive geologic data through Macrostrat API. **When to choose:** Choose this server if you need geological data analysis through natural language queries and are already working with geospatial or earth science data. **When NOT to choose:** Don't choose this if you need real-time geological data updates, as the Macrostrat API may not provide very recent information. ### [AuthorProfileMCP](https://mcpcatalogs.com/en/server/alperenkocyigit-authorprofilemcp) by alperenkocyigit · score 34 · ★2 Academic author network MCP server for analyzing research collaborations and extracting keywords from scholarly profiles. **When to choose:** Choose this server when you need to analyze academic author networks and research collaborations through a well-structured MCP interface with proper rate limiting. **When NOT to choose:** Avoid this server if you need to process large-scale author networks beyond the limits of free-tier APIs, or if you require highly reliable Google Scholar scraping that may occasionally fail due to anti-bot measures. ### [littlesis-mcp](https://mcpcatalogs.com/en/server/code-rabi-littlesis-mcp) by code-rabi · score 32 · ★2 MCP server providing access to LittleSis API for tracking corporate power and relationships. **When to choose:** Choose this server when researching corporate influence networks, political connections, or revolving door dynamics between government and industry, particularly when you need structured data about relationships and accountability. **When NOT to choose:** Don't choose this server if you need real-time data, as LittleSis is updated periodically; or if you require write access to the database, as this server is read-only. --- ## media (47) ### [nuclear](https://mcpcatalogs.com/en/server/nukeop-nuclear) by nukeop · score 60 · ★17,623 Streaming music player that integrates an MCP server for AI control. **When to choose:** Choose Nuclear if you're looking for an open-source music player with AI integration that allows you to control playback through MCP-enabled AI assistants. **When NOT to choose:** Avoid Nuclear if you need commercial music streaming with licensed content or require high-fidelity audio quality, as it sources free music from various sources. ### [XHS-Downloader](https://mcpcatalogs.com/en/server/joeanamier-xhs-downloader) by JoeanAmier · score 59 · ★11,172 XHS-Downloader is a MCP server for downloading content from XiaoHongShu (RedNote) platform with API and MCP capabilities. **When to choose:** Choose XHS-Downloader when you need to programmatically access and download content from Xiaohongshu (RedNote) platform through MCP integration. **When NOT to choose:** Don't choose this if you need access to other social media platforms beyond Xiaohongshu, or if you require advanced filtering and categorization of downloaded content. ### [davinci-resolve-mcp](https://mcpcatalogs.com/en/server/samuelgursky-davinci-resolve-mcp) by samuelgursky · score 56 · ★1,060 A comprehensive MCP server enabling AI assistants to control DaVinci Resolve Studio with 31 compound tools covering 336 API methods. **When to choose:** Choose this MCP server if you need AI assistance to automate DaVinci Resolve workflows, perform color grading, manage media analysis, or integrate video editing into larger AI agent workflows. **When NOT to choose:** Don't choose this if you're using DaVinci Resolve free edition (which doesn't support external scripting) or if you need to modify source media without explicit user confirmation. ### [claude-video-vision](https://mcpcatalogs.com/en/server/jordanrendric-claude-video-vision) by jordanrendric · score 53 · ★613 Claude Code plugin that gives Claude the ability to watch and understand videos through frame extraction and audio analysis. **When to choose:** Choose this MCP server when you need Claude to analyze video content with both visual frames and audio transcription, especially for tutorials, presentations, or recordings where multimodal understanding is essential. **When NOT to choose:** Avoid this if you need high-frequency video processing with proprietary data, as it lacks authentication and may expose API keys; also consider alternatives if you need to process videos in bulk. ### [Gearboy](https://mcpcatalogs.com/en/server/drhelius-gearboy) by drhelius · score 53 · ★1,130 Game Boy emulator with embedded MCP server for AI-assisted debugging and ROM analysis. **When to choose:** Choose this when you need AI-assisted debugging for Game Boy development or ROM hacking, especially if you're already familiar with retro game development tools. **When NOT to choose:** Don't choose this if you need MCP functionality for systems other than Game Boy, or if you're looking for a general-purpose emulator with broader AI capabilities. ### [mcp-youtube-transcript](https://mcpcatalogs.com/en/server/jkawamoto-mcp-youtube-transcript) by jkawamoto · score 50 · ★388 MCP server for retrieving YouTube video transcripts with timestamps and metadata. **When to choose:** Choose this MCP server when you need reliable access to YouTube video transcripts with language support and metadata, particularly for content analysis or accessibility purposes. **When NOT to choose:** Avoid this server if you need offline access to transcripts, as it requires live YouTube API connections and won't work in restricted network environments without proxy configuration. ### [mcp-arr](https://mcpcatalogs.com/en/server/aplaceforallmystuff-mcp-arr) by aplaceforallmystuff · score 50 · ★137 MCP server for unified management of *arr media suite including Sonarr, Radarr, Lidarr, and Prowlarr. **When to choose:** Choose this MCP server if you manage multiple *arr applications and want unified control through natural language queries or a centralized interface. **When NOT to choose:** Avoid if you only use one media service and don't need cross-service functionality, as the server might be overkill for simple setups. ### [short-video-maker](https://mcpcatalogs.com/en/server/gyoridavid-short-video-maker) by gyoridavid · score 50 · ★1,129 An MCP server that creates short-form videos from text using TTS, captions, and background footage. **When to choose:** Choose this server if you need to automate short video creation for social media platforms with minimal editing and want both MCP integration and REST API access. **When NOT to choose:** Don't choose this if you need multilingual support (only English), custom video backgrounds, or have limited system resources (requires ≥3GB RAM). ### [after-effects-mcp](https://mcpcatalogs.com/en/server/dakkshin-after-effects-mcp) by Dakkshin · score 50 · ★358 Adobe After Effects MCP server enabling AI control of compositions, layers, animations via ExtendScript. **When to choose:** Choose this server if you need to integrate AI control with Adobe After Effects for automating creative workflows or building AI-powered video editing tools. **When NOT to choose:** Don't choose this if you need access to newer After Effects features not exposed through ExtendScript, or if you require a solution that works without After Effects installed. ### [unsplash-mcp-server](https://mcpcatalogs.com/en/server/hellokaton-unsplash-mcp-server) by hellokaton · score 49 · ★219 An MCP server for searching Unsplash's high-quality photo library with advanced filters. **When to choose:** Choose this MCP server when you need a simple, efficient way to integrate high-quality image search from Unsplash into your application or workflow. **When NOT to choose:** Avoid if you need commercial use of images beyond Unsplash's license terms, or if you require features beyond search capabilities like image editing or metadata extraction. ### [melrose](https://mcpcatalogs.com/en/server/emicklei-melrose) by emicklei · score 49 · ★208 Melrōse is an interactive programming tool for creating melodies that exposes music composition capabilities through MCP. **When to choose:** Choose Melrōse MCP if you're a musician or developer interested in algorithmic music creation and want real-time interaction with your compositions. **When NOT to choose:** Avoid if you need support for standard audio formats directly or if you prefer traditional DAW interfaces over code-based composition. ### [gimp-mcp](https://mcpcatalogs.com/en/server/maorcc-gimp-mcp) by maorcc · score 49 · ★110 GIMP MCP server lets AI assistants edit images through 56 specialized tools with live visual feedback. **When to choose:** Choose this when you need AI-powered image editing with visual verification, especially for complex multi-step workflows requiring iterative feedback. **When NOT to choose:** Don't choose this if you need non-destructive editing workflows or if you're locked into the Creative Cloud ecosystem. ### [youtube-connector-mcp](https://mcpcatalogs.com/en/server/shellydeng08-youtube-connector-mcp) by ShellyDeng08 · score 48 · ★71 MCP server for YouTube that enables AI assistants to search videos, channels, playlists, and transcripts. **When to choose:** Choose this when you need a dedicated YouTube integration for AI assistants with comprehensive search, content retrieval, and transcript capabilities. **When NOT to choose:** Don't choose this if you need real-time video streaming capabilities or if you're frequently exceeding YouTube API quotas (requires careful quota management). ### [Veo-4-API](https://mcpcatalogs.com/en/server/anil-matcha-veo-4-api) by Anil-matcha · score 48 · ★147 Python wrapper for Veo 4 API with MCP server for AI-generated 4K videos with integrated audio and character consistency. **When to choose:** Choose this MCP server when you need to generate high-quality AI videos with character consistency and integrated audio directly within AI assistants like Claude Desktop or Cursor. **When NOT to choose:** Avoid if you need open-source self-hosted solutions, as this relies on MuAPI's proprietary API with potential vendor lock-in. ### [fal-mcp-server](https://mcpcatalogs.com/en/server/luminarylane-fal-mcp-server) by luminarylane · score 47 · ★44 MCP server for Fal.ai that enables Claude Desktop to generate images, videos, music, and audio using 600+ models. **When to choose:** Choose this MCP server when you need access to a wide variety of media generation models (600+) through Claude Desktop, especially for content creation that requires image, video, or audio generation capabilities. **When NOT to choose:** Avoid this server if you need offline access to models (Fal.ai requires internet), have concerns about vendor lock-in to Fal.ai, or need to process large volumes of media generation requests (free tier has usage limits). ### [fcpxml-mcp-server](https://mcpcatalogs.com/en/server/daredev256-fcpxml-mcp-server) by DareDev256 · score 47 · ★46 AI-powered MCP server for Final Cut Pro XML with 53 tools to analyze, edit, and generate timelines through natural language. **When to choose:** Choose this if you're a Final Cut Pro editor who wants to automate batch operations on timelines using natural language commands, particularly for marker insertion, QC checks, and generating rough cuts. **When NOT to choose:** Don't choose this for real-time editing work or creative decisions that require visual feedback, as it works with exported XML files rather than directly within the editing application. ### [ig-mcp](https://mcpcatalogs.com/en/server/jlbadano-ig-mcp) by jlbadano · score 46 · ★127 Production-ready MCP server enabling AI applications to interact with Instagram Business accounts through the Graph API. **When to choose:** Choose this MCP server if you need to integrate Instagram Business API into AI applications for content management, analytics, or customer interactions through DMs. **When NOT to choose:** Avoid this if you need personal Instagram account integration, don't have a Facebook-connected business account, or require features not available through the Instagram Graph API. ### [Navidrome-MCP](https://mcpcatalogs.com/en/server/blakeem-navidrome-mcp) by Blakeem · score 46 · ★68 MCP server for controlling Navidrome music library with conversation-based playback, discovery tools, and web remote control. **When to choose:** Choose Navidrome-MCP if you use Navidrome as your music server and want conversational control over your music library with advanced playback features and music discovery. **When NOT to choose:** Don't choose this if you don't use Navidrome or prefer a web-based music management interface without local audio playback capabilities. ### [ffmpeg-mcp-lite](https://mcpcatalogs.com/en/server/kevinwatt-ffmpeg-mcp-lite) by kevinwatt · score 46 · ★24 A feature-rich MCP server for video/audio processing using FFmpeg with tools for conversion, compression, trimming, and more. **When to choose:** Choose this server when you need to perform video/audio processing tasks through AI systems without leaving your MCP-compatible environment. **When NOT to choose:** Don't choose this if you need batch processing of many files simultaneously, as the server appears designed for individual file operations through interactive AI agents. ### [cinema4d-mcp](https://mcpcatalogs.com/en/server/ttiimmaacc-cinema4d-mcp) by ttiimmaacc · score 46 · ★83 MCP server connecting Claude AI to Cinema 4D for prompt-driven 3D modeling and scene manipulation. **When to choose:** Choose this server if you're a 3D artist using Cinema 4D who wants to integrate AI assistance into your workflow for faster prototyping and automation of modeling tasks. **When NOT to choose:** Avoid this if you're using other 3D software (it's Cinema 4D-specific), need advanced Redshift material manipulation capabilities, or require stability without Cinema 4D dependencies. ### [ableton-copilot-mcp](https://mcpcatalogs.com/en/server/xiaolaa2-ableton-copilot-mcp) by xiaolaa2 · score 46 · ★78 MCP server for AI assistants to control Ableton Live with song management, track control, MIDI editing, and audio recording capabilities. **When to choose:** Music producers who want to automate repetitive tasks in Ableton Live and integrate AI assistants into their workflow for MIDI editing, track management, and device parameter control. **When NOT to choose:** Users needing write access to Ableton Live features beyond what the ableton-js library supports, or those requiring multi-version compatibility across different Ableton releases. ### [Meting-Agent](https://mcpcatalogs.com/en/server/eldment-meting-agent) by ELDment · score 46 · ★84 Multi-platform music API proxy for AI systems, exposing search, playback, and lyric retrieval via MCP. **When to choose:** Choose Meting-Agent when building AI applications that need to access Chinese music platforms through a unified interface, rather than implementing platform-specific integrations. **When NOT to choose:** Avoid if you need access to non-Chinese music platforms or require advanced music metadata analysis beyond basic search and retrieval functions. ### [photoshop-mcp](https://mcpcatalogs.com/en/server/alisaitteke-photoshop-mcp) by alisaitteke · score 45 · ★44 Photoshop MCP server enables AI assistants to control Adobe Photoshop with 50+ tools for design automation. **When to choose:** Choose this when you need AI-assisted Photoshop automation and are already using or planning to use Claude or Cursor with MCP support. **When NOT to choose:** Don't choose this if you're not an Adobe Photoshop user, or if you need a different image editing software like GIMP or Affinity Photo. ### [claude-homelab](https://mcpcatalogs.com/en/server/jmagar-claude-homelab) by jmagar · score 45 · ★42 HomeLab management plugin for Claude with 16+ integrations for media, infrastructure monitoring, and document management. **When to choose:** Choose this if you're running a complex homelab with media servers, infrastructure monitoring needs, and want a unified interface through Claude. **When NOT to choose:** Don't choose this if you're not using Claude as your primary interface, or if you need a more generic monitoring solution outside the homelab context. ### [mcp-music-studio](https://mcpcatalogs.com/en/server/linxule-mcp-music-studio) by linxule · score 45 · ★45 MCP music studio with ABC notation composition and Strudel live coding, supporting 30+ instruments and interactive sheet music rendering. **When to choose:** Choose this server when you need AI-assisted music composition tools with both notation-based composition and live coding capabilities, especially if you're working within the Claude ecosystem. **When NOT to choose:** Not ideal for professional music production workflows requiring advanced DAW features or integration with existing music production software beyond what this server offers. ### [ableton-mcp](https://mcpcatalogs.com/en/server/jpoindexter-ableton-mcp) by jpoindexter · score 45 · ★22 Ableton Live MCP server with 200+ tools for music production, AI-generated patterns, and full DAW control. **When to choose:** Choose AbletonMCP if you're an electronic music producer or sound designer who wants to leverage AI assistants directly within your Ableton Live workflow, especially if you use Claude or want to integrate with local LLMs via Ollama. **When NOT to choose:** Don't choose this if you primarily work with other DAWs besides Ableton Live, or if you need a general-purpose audio production tool that isn't specifically tied to Ableton's architecture. ### [anilist-mcp](https://mcpcatalogs.com/en/server/yuna0x0-anilist-mcp) by yuna0x0 · score 45 · ★76 AniList MCP server provides comprehensive anime and manga data access through search, retrieval, and management tools. **When to choose:** Choose this server if you need to access or interact with AniList anime and manga data through an MCP interface, especially if you want both local and deployment options. **When NOT to choose:** Don't choose this if you need access to anime data from other platforms like MyAnimeList, or if you require write access to AniList but cannot obtain an API token. ### [mcp-video](https://mcpcatalogs.com/en/server/kyanitelabs-mcp-video) by KyaniteLabs · score 44 · ★18 Video editing MCP server with 103 tools for FFmpeg operations, AI-assisted media processing, and local repurposing. **When to choose:** Choose this when you need extensive video editing capabilities for AI agents and want to avoid brittle shell commands with structured tool schemas. **When NOT to choose:** Avoid if you need cloud-based video processing, as this server requires local FFmpeg installation and is limited to local-only operations. ### [mcp-claude-spotify](https://mcpcatalogs.com/en/server/imprvhub-mcp-claude-spotify) by imprvhub · score 44 · ★35 MCP server for Spotify integration with Claude Desktop, providing music control and playlist management capabilities. **When to choose:** Choose this MCP server if you want Claude Desktop to control your Spotify account through natural language commands and need full playlist management capabilities. **When NOT to choose:** Avoid if you're not using Claude Desktop or if you need to integrate with other AI assistants that don't support MCP. ### [video-research-mcp](https://mcpcatalogs.com/en/server/galbaz1-video-research-mcp) by Galbaz1 · score 43 · ★21 MCP server providing 51 video analysis and research tools through Gemini 3.1 Pro with Weaviate integration. **When to choose:** Choose this if you need video analysis, deep research capabilities, and media production tools in one MCP server, especially if you're already using Google's Gemini API. **When NOT to choose:** Avoid if you need multi-provider AI model support beyond Gemini, or if you're concerned about vendor lock-in with Google's ecosystem. ### [mcp-koii](https://mcpcatalogs.com/en/server/benjaminr-mcp-koii) by benjaminr · score 43 · ★23 MCP server for controlling Teenage Engineering EP-133 K.O. II sampler via MIDI with pattern creation and playback. **When to choose:** Choose this MCP server if you own a Teenage Engineering EP-133 K.O. II and want to integrate it with AI assistants like Claude for music creation and control. **When NOT to choose:** Don't choose this if you don't own the specific EP-133 hardware, as it's designed exclusively for this particular device. ### [mcp-youtube](https://mcpcatalogs.com/en/server/adhikasp-mcp-youtube) by adhikasp · score 41 · ★48 A Python-based MCP server that fetches YouTube transcripts with timestamps for LLM integration. **When to choose:** Choose this when you need to integrate YouTube video transcripts into your AI workflow and want a simple, dedicated MCP solution for accessing YouTube content. **When NOT to choose:** Don't choose this if you need to process audio/video directly, handle YouTube API rate limits, or require sophisticated YouTube content analysis beyond transcript retrieval. ### [mcp-server-giphy](https://mcpcatalogs.com/en/server/magarcia-mcp-server-giphy) by magarcia · score 41 · ★32 MCP server integrating Giphy API for AI models to search, retrieve, and utilize GIFs. **When to choose:** Choose this server when building AI applications that need to incorporate GIFs as responses or content, especially if you're already using TypeScript in your development stack. **When NOT to choose:** Don't choose this server if you need extensive customization beyond the provided search methods, or if you require offline access to GIF content. ### [mcp-kling](https://mcpcatalogs.com/en/server/199-mcp-mcp-kling) by 199-mcp · score 41 · ★26 MCP Kling is a comprehensive server providing full access to Kling AI's video generation tools via MCP protocol. **When to choose:** Choose this MCP server when you need comprehensive video generation capabilities through Claude and want automatic download of all generated content without manual file management. **When NOT to choose:** Avoid if you need video generation beyond 10 seconds, require commercial licensing for generated content, or have concerns about vendor lock-in with Kling's API. ### [mcp_newsnow](https://mcpcatalogs.com/en/server/sligter-mcp-newsnow) by sligter · score 40 · ★69 A news aggregation MCP server providing trending topics from 14+ Chinese and English platforms. **When to choose:** Choose this server when you need real-time access to trending news across Chinese platforms and want to integrate it into MCP-compatible AI workflows. **When NOT to choose:** Avoid this server if you need news sources outside of China or require more advanced filtering capabilities beyond basic source selection. ### [mcp-spotify-player](https://mcpcatalogs.com/en/server/vsaez-mcp-spotify-player) by vsaez · score 40 · ★19 An MCP server that integrates Spotify control with AI assistants for natural music commands. **When to choose:** Choose this if you want to control Spotify directly through your AI assistant without switching apps, especially if you're already using Claude Desktop or JetBrains IDEs with GitHub Copilot. **When NOT to choose:** Don't choose this if you don't have Spotify Premium, if you're using a different AI assistant that doesn't support MCP, or if you prefer a dedicated music player interface over command-line control. ### [youtube-mcp-server](https://mcpcatalogs.com/en/server/0gis0-youtube-mcp-server) by 0GiS0 · score 40 · ★17 A practical MCP server that integrates YouTube API with both stdio and SSE transport methods. **When to choose:** Choose this MCP server if you need to integrate YouTube data into your AI workflows and want both local and remote connection options. **When NOT to choose:** Don't choose this if you need full YouTube functionality beyond channel search and information retrieval, or if you prefer a simpler installation process. ### [mcp-server](https://mcpcatalogs.com/en/server/plainly-videos-mcp-server) by plainly-videos · score 39 · ★6 Official MCP server for Plainly Videos API enabling After Effects automation through LLM clients. **When to choose:** Choose this if you're already using or planning to use Plainly Videos for video automation and want to integrate it with LLM agents. **When NOT to choose:** Don't choose this if you need video automation capabilities for other platforms or if you prefer not to rely on a single vendor's API. ### [youtube-transcript-mcp](https://mcpcatalogs.com/en/server/lucasliet-youtube-transcript-mcp) by lucasliet · score 39 · ★4 MCP server for extracting YouTube video transcripts with timestamps. **When to choose:** Choose this tool when you need to process YouTube video transcripts within an AI agent workflow and want both local and remote deployment options. **When NOT to choose:** Don't choose this if you need to process videos from platforms other than YouTube, or if you require transcript caching or persistent storage functionality. ### [json2video-mcp-server](https://mcpcatalogs.com/en/server/omergocmen-json2video-mcp-server) by omergocmen · score 38 · ★25 An MCP server for generating videos programmatically using the json2video API with extensive scene and element support. **When to choose:** Choose this when you need to integrate video generation capabilities into your LLM workflows without building custom video processing infrastructure. **When NOT to choose:** Don't choose this if you need real-time video generation, require advanced video editing features, or want to avoid cloud-based video processing services. ### [videocapture-mcp](https://mcpcatalogs.com/en/server/13rac1-videocapture-mcp) by 13rac1 · score 37 · ★17 Python MCP server for accessing and controlling webcams via OpenCV to capture images and manipulate camera settings. **When to choose:** Choose this server when you need AI agents to capture static images from webcams for visual analysis or document processing. **When NOT to choose:** Avoid this server if you need real-time video streaming capabilities or advanced image processing beyond basic capture. ### [MCP2OSC](https://mcpcatalogs.com/en/server/yyf-mcp2osc) by yyf · score 36 · ★3 MCP2OSC enables parametric control of OSC devices using natural language through Claude. **When to choose:** Choose MCP2OSC when working with creative coding environments like MaxMSP, PureData, or SuperCollider and need intuitive natural language control over OSC parameters without manual knob adjustments. **When NOT to choose:** Don't choose MCP2OSC if you need broader control beyond OSC protocol or if you're working with multimedia systems that don't support OSC communication. ### [omdb-mcp-server](https://mcpcatalogs.com/en/server/tyrell-omdb-mcp-server) by tyrell · score 34 · ★1 A Java-based MCP server for OMDB API that enables AI assistants to search for movies and retrieve detailed movie information. **When to choose:** Choose this server when you need movie data integration for AI assistants and want the performance benefits of intelligent caching. **When NOT to choose:** Don't choose this if you need real-time movie data (cache has 1-hour TTL) or want a language other than Java implementation. ### [podbeanMCP](https://mcpcatalogs.com/en/server/amurshak-podbeanmcp) by amurshak · score 33 · ★0 MCP server for managing podcasts through the Podbean API with authentication, episodes, and analytics. **When to choose:** Choose this server if you're already using Podbean for podcast hosting and want to integrate podcast management directly into your AI workflow. **When NOT to choose:** Don't choose this server if you need direct file upload capabilities or if you're using a different podcast hosting platform. ### [japan-ani-search-mcp](https://mcpcatalogs.com/en/server/hirofumitorato-japan-ani-search-mcp) by hirofumitorato · score 33 · ★0 MCP server for searching anime and manga using AniList API with tools for character details and cultural explanations. **When to choose:** Choose this server when you need comprehensive anime and manga information within AI workflows, especially for cultural context explanations or seasonal content discovery. **When NOT to choose:** Don't choose this if you need manga-specific features beyond what AniList provides, or if you require real-time updates as the API is third-party. ### [ffmpeg-mcp-lite](https://mcpcatalogs.com/en/server/jasmyname-ffmpeg-mcp-lite) by JasMyName · score 31 · ★1 An MCP server that wraps FFmpeg capabilities for video and audio processing through AI commands. **When to choose:** When you need a simple GUI for basic FFmpeg operations and are comfortable using standalone applications rather than MCP integrations. **When NOT to choose:** When you need actual MCP server integration with AI assistants, as this appears to be a GUI wrapper without protocol implementation. ### [image-processing-mcp-server](https://mcpcatalogs.com/en/server/rafael-castelo-image-processing-mcp-server) by rafael-castelo · score 28 · ★1 MCP server for image processing tasks including resizing, compression, format conversion, metadata extraction, and cropping. **When to choose:** Choose this MCP server when you need to perform image processing tasks directly within your AI coding assistant without leaving your IDE. **When NOT to choose:** Avoid this server for production-grade image processing pipelines or applications requiring advanced features like OCR, object detection, or machine learning-based image analysis. --- ## monitoring (21) ### [PerformanceMonitor](https://mcpcatalogs.com/en/server/erikdarlingdata-performancemonitor) by erikdarlingdata · score 53 · ★380 SQL Server performance monitoring tool with built-in MCP server for AI-powered analysis of wait stats and query performance. **When to choose:** Choose this if you need AI-powered SQL Server performance analysis without third-party tool subscriptions and prefer keeping all data local. **When NOT to choose:** Not suitable if you need multi-database monitoring across different database types or require write access capabilities from your AI tools. ### [prometheus-mcp-server](https://mcpcatalogs.com/en/server/pab1it0-prometheus-mcp-server) by pab1it0 · score 52 · ★442 Prometheus MCP server enables AI agents to query and analyze metrics through standardized interfaces. **When to choose:** Choose this server when you need AI agents to query Prometheus metrics and you prefer a containerized, production-ready solution with multiple deployment options. **When NOT to choose:** Don't choose this if you need write access to your metrics or are using a monitoring solution other than Prometheus. ### [loki-mcp](https://mcpcatalogs.com/en/server/grafana-loki-mcp) by grafana · score 51 · ★141 A production-ready MCP server for querying Grafana Loki logs with comprehensive documentation and multiple deployment options. **When to choose:** Choose this server when you need to query Grafana Loki logs through natural language interfaces and already have a Loki deployment. **When NOT to choose:** Don't choose this if you need to query other logging systems besides Grafana Loki, or if you require advanced log processing capabilities beyond simple LogQL queries. ### [mcp-victoriametrics](https://mcpcatalogs.com/en/server/victoriametrics-mcp-victoriametrics) by VictoriaMetrics · score 50 · ★167 MCP server for VictoriaMetrics providing full API access to monitoring and observability data. **When to choose:** Choose this server if you're already using VictoriaMetrics for monitoring and want AI-powered access to query metrics, analyze alerting rules, and debug monitoring systems through MCP integration. **When NOT to choose:** Don't choose this if you need write access to your metrics data or are using a different monitoring system that doesn't integrate with VictoriaMetrics. ### [jetski](https://mcpcatalogs.com/en/server/hyprmcp-jetski) by hyprmcp · score 50 · ★209 An MCP analytics and authentication platform with zero code changes that provides OAuth2.1, real-time logs, and client onboarding. **When to choose:** Choose Jetski when you need to add authentication and analytics to existing MCP servers without modifying them, and when you want visibility into prompt-tool usage patterns. **When NOT to choose:** Don't choose Jetski if you need fine-grained control over authentication logic beyond OAuth2.1, or if you're building a simple MCP server without advanced monitoring needs. ### [unraid-mcp](https://mcpcatalogs.com/en/server/jmagar-unraid-mcp) by jmagar · score 48 · ★64 A comprehensive MCP server for Unraid server management through GraphQL API, supporting system info, Docker, VMs, array operations and live telemetry. **When to choose:** Choose this MCP server if you need comprehensive programmatic control over Unraid NAS systems, particularly for automation of maintenance tasks or integration with existing monitoring workflows. **When NOT to choose:** Avoid this if you don't use Unraid systems, need broader NAS management capabilities, or require integration with non-MCP tooling that doesn't support the protocol. ### [zabbix-mcp-server](https://mcpcatalogs.com/en/server/initmax-zabbix-mcp-server) by initMAX · score 48 · ★98 Complete Zabbix API server exposing 237 tools for monitoring and management via MCP protocol. **When to choose:** Choose this if you need AI integration with an existing Zabbix monitoring infrastructure and want comprehensive API access with advanced features like PDF reports and anomaly detection. **When NOT to choose:** Don't choose this if you're not using Zabbix as your monitoring system, as this server only provides access to Zabbix APIs. ### [ingero](https://mcpcatalogs.com/en/server/ingero-io-ingero) by ingero-io · score 48 · ★81 Ingero is an eBPF-based GPU observability tool that exposes diagnostic capabilities through MCP for AI assistants. **When to choose:** Choose Ingero when you need deep causal analysis of GPU performance issues in ML/AI workloads and want to query insights through AI assistants without SQL knowledge. **When NOT to choose:** Don't choose Ingero if you need cross-platform support (Linux-only), require very low overhead (<0.5%), or need to analyze non-GPU performance issues. ### [mcp-victorialogs](https://mcpcatalogs.com/en/server/victoriametrics-mcp-victorialogs) by VictoriaMetrics · score 48 · ★73 VictoriaLogs MCP server provides access to VictoriaLogs observability data through Model Context Protocol. **When to choose:** Choose this when you need AI-driven access to VictoriaLogs for log analysis, metrics, and observability without writing custom API clients. **When NOT to choose:** Don't choose this if you need write access to VictoriaLogs or require integration with non-Victoria observability stacks. ### [signoz-mcp-server](https://mcpcatalogs.com/en/server/signoz-signoz-mcp-server) by SigNoz · score 47 · ★92 MCP server providing seamless access to SigNoz observability data through AI assistants and LLMs. **When to choose:** Choose this if you're already using SigNoz for observability and want to enable AI-powered querying of your metrics, traces, logs, alerts, and dashboards through natural language. **When NOT to choose:** Don't choose this if you're not using SigNoz for observability, as it requires a SigNoz instance and won't work with other monitoring systems. ### [cosmotop](https://mcpcatalogs.com/en/server/bjia56-cosmotop) by bjia56 · score 46 · ★67 Multiplatform system monitoring tool with MCP server mode exposing system metrics via Model Context Protocol. **When to choose:** Choose cosmotop when you need cross-platform system monitoring through MCP and value the single-executable distribution model that works across Linux, Windows, macOS, and BSD systems. **When NOT to choose:** Avoid cosmotop if you need deeply specialized monitoring metrics beyond its provided tools or if you require authentication for MCP server access, which isn't implemented. ### [mcp-monitor](https://mcpcatalogs.com/en/server/seekrays-mcp-monitor) by seekrays · score 43 · ★82 Go-based MCP server exposing system metrics including CPU, memory, disk, network, and process information. **When to choose:** Choose this MCP server when you need basic system metrics monitoring through an MCP interface without complex setup, especially for server administration tasks. **When NOT to choose:** Don't choose this if you need advanced monitoring with historical data, alerting, or if you require authentication/authorization features as it lacks these capabilities. ### [prometheus-mcp](https://mcpcatalogs.com/en/server/idanfishman-prometheus-mcp) by idanfishman · score 43 · ★25 MCP server enabling AI agents to query Prometheus monitoring metrics through natural language commands. **When to choose:** Choose this when you need AI agents to analyze Prometheus monitoring data through natural language queries, especially if you're already using Prometheus and want to enhance it with AI capabilities without additional infrastructure. **When NOT to choose:** Don't choose this if you need write access to your monitoring system or if you're using a monitoring solution other than Prometheus that doesn't expose a compatible API. ### [appcontrol-mcp-go](https://mcpcatalogs.com/en/server/appcontrollabs-appcontrol-mcp-go) by AppControlLabs · score 43 · ★22 AppControl MCP server provides read-only access to historical resource usage and system security data through natural language queries. **When to choose:** Choose AppControl MCP when you need deep historical system monitoring and security insights for Windows PCs and want to query this data through natural language to AI assistants. **When NOT to choose:** Avoid this MCP server if you need cross-platform support (Windows-only) or write access to system settings (this server is read-only only). ### [grafana-mcp-analyzer](https://mcpcatalogs.com/en/server/sailingcoder-grafana-mcp-analyzer) by SailingCoder · score 42 · ★43 An MCP server that enables AI assistants to directly analyze Grafana monitoring data through natural language queries. **When to choose:** Choose this server if your team already uses Grafana for monitoring and wants to leverage AI assistants for direct natural language analysis of monitoring data without additional data processing. **When NOT to choose:** Avoid this server if you need write access to monitoring data (this server is read-only) or if you require support for monitoring systems other than Grafana. ### [mcp-status-observer](https://mcpcatalogs.com/en/server/imprvhub-mcp-status-observer) by imprvhub · score 41 · ★7 MCP server for monitoring operational status of major digital platforms like AI providers, cloud services, and developer tools. **When to choose:** Choose this MCP server if you need real-time status monitoring of multiple digital platforms and developer tools in one place, especially when your workflows depend on the availability of AI services, cloud infrastructure, and collaboration tools. **When NOT to choose:** Not suitable if you need to monitor internal/private services, require custom alerting mechanisms, or need historical status analysis beyond recent incidents - this tool focuses on public APIs of major platforms. ### [chronolog](https://mcpcatalogs.com/en/server/grc-iit-chronolog) by grc-iit · score 38 · ★9 ChronoLog MCP server enables real-time logging and event processing with LLMs through a distributed time-structured log storage system. **When to choose:** Choose ChronoLog when you need high-performance distributed log storage with time-based data ordering and real-time LLM integration for enterprise logging and event processing. **When NOT to choose:** Avoid ChronoLog if you need simple logging solutions, as it's designed for complex distributed environments and may be overkill for basic use cases. ### [hormuz-strait-mcp](https://mcpcatalogs.com/en/server/mcp-use-hormuz-strait-mcp) by mcp-use · score 37 · ★1 MCP server providing tactical maritime situation dashboard for Strait of Hormuz with interactive maps and ship tracking tools. **When to choose:** Choose Hormuz Watch for specialized maritime situational awareness of the Strait of Hormuz with visual map interfaces and real-time AIS tracking capabilities. **When NOT to choose:** Don't choose if you need maritime data for regions outside the Strait of Hormuz, or if you require a commercial-grade AIS API (the current VesselFinder integration may be rate-limited). ### [prometheus_mcp_server](https://mcpcatalogs.com/en/server/caesaryangs-prometheus-mcp-server) by CaesarYangs · score 35 · ★34 An MCP server for LLMs to query, analyze and interact with Prometheus databases. **When to choose:** Choose this server when you need to integrate Prometheus monitoring data with LLM workflows, especially when you want to analyze system metrics through natural language queries without writing PromQL. **When NOT to choose:** Don't choose this if you need to write to Prometheus (this server is read-only only), or if your Prometheus instance requires complex authentication beyond basic host configuration. ### [mcp-server-prometheus](https://mcpcatalogs.com/en/server/loglmhq-mcp-server-prometheus) by loglmhq · score 34 · ★17 MCP server that bridges Claude with Prometheus metrics and data. **When to choose:** Choose this for Prometheus integration if you need basic metric access and metadata retrieval through MCP. **When NOT to choose:** Avoid if you need complex metric querying, time series analysis, or advanced Prometheus features not yet implemented. ### [mcp-ayd-server](https://mcpcatalogs.com/en/server/macrat-mcp-ayd-server) by macrat · score 27 · ★0 MCP server for monitoring Ayd status checks via the Model Context Protocol. **When to choose:** Choose this MCP server if you use Ayd for service monitoring and want to integrate status checks into your AI workflows. **When NOT to choose:** Don't choose this if you don't use Ayd or need more comprehensive monitoring features beyond basic status checking. --- ## notion (4) ### [mcp-notion-server](https://mcpcatalogs.com/en/server/suekou-mcp-notion-server) by suekou · score 55 · ★892 MCP server for connecting Notion to MCP-compatible clients, enabling AI agents to find, read, query and update Notion workspaces. **When to choose:** Choose this server for comprehensive Notion integration with AI agents when you need both high-level tools for common operations and low-level access to the full Notion API. **When NOT to choose:** Avoid if you need read-only access (this server requires write permissions) or if you're concerned about vendor lock-in to Notion's API. ### [better-notion-mcp](https://mcpcatalogs.com/en/server/n24q02m-better-notion-mcp) by n24q02m · score 46 · ★28 A Markdown-first Notion API server providing 10 composite tools to simplify Notion integration for AI agents. **When to choose:** Choose this when you need a high-level abstraction over Notion API operations with better readability through Markdown and reduced token usage for AI agents. **When NOT to choose:** Avoid if you need direct access to all Notion API features not covered by the composite tools, or if you require complex permission management beyond the provided OAuth model. ### [notion-mcp-server](https://mcpcatalogs.com/en/server/awkoy-notion-mcp-server) by awkoy · score 44 · ★151 Production-ready MCP server for Notion API integration with comprehensive tools for managing pages, databases, comments, and users. **When to choose:** Choose this server if you need AI assistants to interact directly with Notion's API for content creation, management, and knowledge retrieval. **When NOT to choose:** Don't choose if you need access to other productivity tools beyond Notion, or if you require write permissions to Notion workspaces that the integration can't access. ### [notion-export-mcp-server](https://mcpcatalogs.com/en/server/syucream-notion-export-mcp-server) by syucream · score 28 · ★1 An MCP server that exports Notion pages to markdown via unofficial API. **When to choose:** Choose this when you need simple, one-way export of Notion pages to markdown for AI analysis or processing. **When NOT to choose:** Avoid if you need write access to Notion, or if you're concerned about API stability as it relies on Notion's unofficial internal API. --- ## ops-infra (35) ### [radar](https://mcpcatalogs.com/en/server/skyhook-io-radar) by skyhook-io · score 56 · ★1,934 Radar is a Kubernetes UI with built-in MCP server for AI assistants to query cluster resources. **When to choose:** Choose Radar when you need to provide AI assistants with access to your Kubernetes cluster data for troubleshooting, analysis, and monitoring purposes without exposing sensitive credentials. **When NOT to choose:** Avoid Radar if you need deep integration with cloud provider APIs or require a fully cloud-managed solution without local components. ### [kubernetes-mcp-server](https://mcpcatalogs.com/en/server/containers-kubernetes-mcp-server) by containers · score 55 · ★1,593 Go-based native MCP server for Kubernetes and OpenShift with direct API integration, supporting CRUD operations, pods, namespaces, Helm, and Tekton. **When to choose:** Choose this for direct Kubernetes API access with high performance and multi-cluster support, especially when you need lightweight, native binaries without external dependencies. **When NOT to choose:** Avoid if you need extensive third-party integrations beyond Kubernetes core resources, or if you're already invested in a different MCP server ecosystem with established tooling. ### [kubectl-mcp-server](https://mcpcatalogs.com/en/server/rohitg00-kubectl-mcp-server) by rohitg00 · score 54 · ★888 A comprehensive MCP server for Kubernetes infrastructure management through natural language. **When to choose:** Choose this when you need to manage Kubernetes through natural language and already use an AI assistant like Claude or Cursor for development work. **When NOT to choose:** Avoid if you need fine-grained control over kubectl commands or require features beyond what MCP protocol can expose. ### [Unla](https://mcpcatalogs.com/en/server/amoylab-unla) by AmoyLab · score 54 · ★2,113 Unla is a lightweight gateway service that transforms existing MCP Servers and APIs into MCP endpoints without code changes. **When to choose:** Choose Unla when you need to integrate existing APIs or MCP servers with MCP clients without code modifications, especially when you need multi-tenant support and a management UI. **When NOT to choose:** Avoid Unla if you require low-level customization of the MCP protocol or if you need support for protocols that are not yet implemented (like gRPC or WebSocket conversions). ### [esp32_nat_router](https://mcpcatalogs.com/en/server/martin-ger-esp32-nat-router) by martin-ger · score 54 · ★1,956 ESP32 NAT Router firmware with MCP Bridge for AI-controlled networking features. **When to choose:** Choose this if you need a low-cost network router with AI control capabilities, especially for IoT environments or small networks, and want to leverage the ESP32's flexibility with MCP integration. **When NOT to choose:** Don't choose this for high-performance networks, as it has limited throughput (5-15 Mbps) and is not intended as a replacement for professional home routers. ### [unifi-mcp](https://mcpcatalogs.com/en/server/sirkirby-unifi-mcp) by sirkirby · score 53 · ★339 MCP servers for UniFi applications (Network, Protect, Access) with 169 tools total and cross-product correlations. **When to choose:** Choose UniFi MCP if you're already using Ubiquiti UniFi infrastructure and want AI-powered management capabilities with advanced features like preview-before-confirm mutations and cross-product event correlation. **When NOT to choose:** Don't choose UniFi MCP if you're not using Ubiquiti hardware, as this server only supports UniFi controllers and doesn't integrate with other networking ecosystems. ### [cloud-run-mcp](https://mcpcatalogs.com/en/server/googlecloudplatform-cloud-run-mcp) by GoogleCloudPlatform · score 53 · ★608 MCP server that enables AI agents to deploy applications to Google Cloud Run with comprehensive tool support and authentication options. **When to choose:** Choose this when you're already using Google Cloud and want AI assistants to manage Cloud Run deployments without switching contexts. **When NOT to choose:** Don't choose this if you're not using Google Cloud, or if you need deployment features beyond what Cloud Run offers. ### [coolify-mcp](https://mcpcatalogs.com/en/server/stumason-coolify-mcp) by StuMason · score 53 · ★372 Comprehensive MCP server with 38 token-optimized tools for managing Coolify self-hosted PaaS through AI assistants. **When to choose:** Choose this if you're using Coolify for self-hosted applications and want AI assistants to manage and debug your infrastructure through natural language commands with drastically reduced token usage. **When NOT to choose:** Don't choose this if you're not using Coolify, as it's specifically designed to interact only with the Coolify API. ### [tfmcp](https://mcpcatalogs.com/en/server/nwiizo-tfmcp) by nwiizo · score 52 · ★364 tfmcp is a comprehensive Terraform MCP server enabling AI assistants to manage infrastructure with 31 specialized tools. **When to choose:** Choose tfmcp when you need comprehensive Terraform management through AI assistants, especially if you're already using Claude Desktop and require enterprise-grade security features. **When NOT to choose:** Avoid tfmcp if you need MCP server support for cloud providers other than Terraform, or if you require write capabilities beyond Terraform operations. ### [mcpcan](https://mcpcatalogs.com/en/server/kymo-mcp-mcpcan) by Kymo-MCP · score 52 · ★719 MCPCAN is a centralized management platform for MCP services with container deployment, monitoring, and security features. **When to choose:** Choose MCPCan when you need centralized management of multiple MCP services with container deployment, security features, and protocol conversion capabilities. **When NOT to choose:** Avoid MCPCan if you prefer simpler MCP deployment without containerization or need a lightweight solution for just one MCP service. ### [mcp-ssh-manager](https://mcpcatalogs.com/en/server/bvisible-mcp-ssh-manager) by bvisible · score 51 · ★203 A comprehensive MCP server providing 37 tools for remote SSH management, DevOps automation, backups, database operations, and health monitoring. **When to choose:** Choose MCP SSH Manager when you need comprehensive server management capabilities through an AI interface, especially for DevOps automation, monitoring, and database operations. **When NOT to choose:** Avoid if you need Windows-specific SSH features not supported by OpenSSH, or if you require more specialized cloud provider integration beyond SSH. ### [k8s-mcp-server](https://mcpcatalogs.com/en/server/alexei-led-k8s-mcp-server) by alexei-led · score 51 · ★209 K8s MCP Server enables Claude to securely execute Kubernetes commands via MCP protocol. **When to choose:** Choose this server if you need Claude to interact with Kubernetes clusters securely, especially when you want to leverage AI assistance for cluster management without exposing credentials directly to AI. **When NOT to choose:** Avoid if you need support for other container orchestration platforms like Docker Swarm or Nomad, or if you require fine-grained access control beyond what kubectl provides. ### [aws-mcp-server](https://mcpcatalogs.com/en/server/alexei-led-aws-mcp-server) by alexei-led · score 50 · ★182 AWS MCP server enables AI assistants to execute AWS CLI commands through MCP in safe containerized environment. **When to choose:** Choose this when you need Claude or other MCP-aware AI assistants to have broad but controlled access to AWS infrastructure via command line, especially with existing AWS CLI knowledge. **When NOT to choose:** Avoid if you need more granular access control beyond IAM policies, or if you prefer direct AWS SDK integration rather than CLI-based operations. ### [boltmcp](https://mcpcatalogs.com/en/server/boltmcp-boltmcp) by boltmcp · score 50 · ★350 BoltMCP is an enterprise-grade platform for creating and managing custom MCP servers on-premises with OAuth 2 security. **When to choose:** Choose BoltMCP when you need enterprise-grade on-premises MCP orchestration with OAuth 2 authentication and the ability to create highly customized MCP servers with composed API operations. **When NOT to choose:** Don't choose BoltMCP if you need a simple, lightweight MCP solution or if you're looking for a cloud-hosted service rather than on-premises infrastructure. ### [ProxmoxMCP-Plus](https://mcpcatalogs.com/en/server/rekklesna-proxmoxmcp-plus) by RekklesNA · score 50 · ★191 MCP server enabling Proxmox VE VM and LXC control through MCP and OpenAPI interfaces. **When to choose:** Choose this when you need unified control of Proxmox environments through both LLM/AI agents and traditional automation tools, with emphasis on safety features and persistent job tracking. **When NOT to choose:** Avoid if you're not using Proxmox VE or need more specialized cloud infrastructure management tools with broader capabilities. ### [spotinfo](https://mcpcatalogs.com/en/server/alexei-led-spotinfo) by alexei-led · score 50 · ★161 MCP server for exploring AWS EC2 Spot instance inventory with real-time placement scores, pricing data, and interruption rates. **When to choose:** Choose this server when you need to explore AWS spot instance options with placement scores and compare pricing across regions through AI assistants. **When NOT to choose:** Don't choose if you need spot instance management capabilities beyond exploration, or if you're working with cloud providers other than AWS. ### [mcp-k8s](https://mcpcatalogs.com/en/server/silenceper-mcp-k8s) by silenceper · score 49 · ★145 Go-based MCP server for Kubernetes and Helm operations with granular control permissions. **When to choose:** Choose this MCP server if you need comprehensive Kubernetes and Helm operations through MCP with fine-grained control over write permissions. **When NOT to choose:** Avoid if you need a multi-cluster management solution with advanced scheduling capabilities, as this focuses on single-cluster operations. ### [mcp-server](https://mcpcatalogs.com/en/server/octopusdeploy-mcp-server) by OctopusDeploy · score 48 · ★96 Official Octopus Deploy MCP server enabling AI assistants to interact with DevOps workflows and deployments. **When to choose:** Choose this if you're already using Octopus Deploy for deployment automation and want to integrate AI assistants into your DevOps workflows. **When NOT to choose:** Avoid this if you're not using Octopus Deploy, as it's vendor-specific and requires an active Octopus instance. ### [ksail](https://mcpcatalogs.com/en/server/devantler-tech-ksail) by devantler-tech · score 48 · ★147 KSail is a comprehensive Kubernetes SDK that operates as an MCP server, providing cluster management across multiple distributions with built-in GitOps, secrets management, and AI integration. **When to choose:** Choose KSail if you need a unified Kubernetes management tool across multiple distributions and providers with integrated AI assistance through its MCP server. **When NOT to choose:** Avoid KSail if you need advanced enterprise features like sophisticated RBAC or if its PolyForm Shield License's non-compete clause restricts your use case. ### [aks-mcp](https://mcpcatalogs.com/en/server/azure-aks-mcp) by Azure · score 48 · ★132 Azure's AKS-MCP server enables AI assistants to interact with Azure Kubernetes Service clusters through MCP protocol. **When to choose:** Choose AKS-MCP when you need to interact with Azure Kubernetes Service through AI assistants and want direct CLI integration with Azure resources. **When NOT to choose:** Avoid if you're not using Azure AKS, or if you require authentication methods beyond Azure CLI integration. ### [homebutler](https://mcpcatalogs.com/en/server/higangssh-homebutler) by Higangssh · score 48 · ★151 HomeButler is a single-binary homelab management tool with MCP server capabilities for monitoring and managing servers. **When to choose:** Choose HomeButler if you're a homelab user or small operations team wanting AI-safe server monitoring without exposing SSH credentials. **When NOT to choose:** Don't choose this for large enterprise deployments with complex authentication requirements or if you need write access to server configurations. ### [mikrotik-mcp](https://mcpcatalogs.com/en/server/jeff-nasseri-mikrotik-mcp) by jeff-nasseri · score 48 · ★171 MikroTik MCP server enables AI assistants to manage RouterOS devices through natural language commands. **When to choose:** Choose this MCP server if you need AI-driven management of MikroTik network devices and want to automate router administration through natural language commands. **When NOT to choose:** Don't choose this if you're not using MikroTik devices, or need support for other router brands like Cisco or Juniper. ### [mcp-gateway](https://mcpcatalogs.com/en/server/theognis1002-mcp-gateway) by theognis1002 · score 47 · ★41 Production-ready API gateway for MCP servers with authentication, logging, rate limiting, and multi-protocol support. **When to choose:** Choose MCP Gateway when you need enterprise-grade security and multi-protocol support for MCP servers in a production environment, especially when managing multiple MCP servers across different namespaces. **When NOT to choose:** Avoid for simple single-server deployments or when you need lightweight resource usage, as this is a full-featured gateway with significant overhead. ### [peta-core](https://mcpcatalogs.com/en/server/dunialabs-peta-core) by dunialabs · score 47 · ★47 Production-ready MCP control plane with security, policy enforcement, and audit capabilities. **When to choose:** Choose Peta Core when you need a production-ready control plane for MCP with enterprise-grade security, centralized policy management, and comprehensive audit capabilities for multiple downstream servers. **When NOT to choose:** Avoid Peta Core if you need a self-contained solution without database dependencies, prefer simpler authentication models, or are unwilling to accept the Elastic License 2.0 restrictions on hosting as a managed service. ### [podman-mcp-server](https://mcpcatalogs.com/en/server/manusa-podman-mcp-server) by manusa · score 46 · ★73 A well-documented MCP server for managing Podman and Docker containers with comprehensive tool coverage. **When to choose:** Choose this if you're working with Podman or Docker containers and want to manage them through AI assistants or MCP-enabled applications. **When NOT to choose:** Don't choose this if you're using container runtimes other than Podman/Docker, or need advanced orchestration features not covered by the available tools. ### [mcp-kubernetes](https://mcpcatalogs.com/en/server/azure-mcp-kubernetes) by Azure · score 46 · ★57 An MCP server enabling AI assistants to interact with Kubernetes clusters through kubectl commands. **When to choose:** Choose this for AI assistants that need to interact with Kubernetes clusters, especially when you want granular control over access levels and a unified tool interface to minimize context usage. **When NOT to choose:** Don't use this if you need to integrate with other cloud providers beyond Kubernetes, or if your AI tools require direct API access to Kubernetes rather than kubectl command translation. ### [mcp-1panel](https://mcpcatalogs.com/en/server/1panel-dev-mcp-1panel) by 1Panel-dev · score 45 · ★147 MCP server for 1Panel, providing tools for website, database, and server management. **When to choose:** Choose this MCP server if you're already using 1Panel for server management and want to enable AI assistants to handle routine server tasks like website creation, database management, and system monitoring. **When NOT to choose:** Don't choose this if you're not using 1Panel or prefer a vendor-agnostic approach to server management, as this server is tightly coupled with 1Panel's ecosystem. ### [arch-mcp](https://mcpcatalogs.com/en/server/nihalxkumar-arch-mcp) by nihalxkumar · score 45 · ★45 A comprehensive MCP server providing AI access to Arch Linux ecosystems including Wiki, AUR, and repositories. **When to choose:** Choose this server if you're working with Arch Linux systems and want AI assistance with package management, system diagnostics, or accessing Arch Wiki/AUR information. **When NOT to choose:** Avoid this server if you're not using Arch Linux, as many of its tools are specific to the Arch ecosystem and won't work on other distributions. ### [commvault-mcp-server](https://mcpcatalogs.com/en/server/commvault-commvault-mcp-server) by Commvault · score 44 · ★13 Commvault MCP Server enables AI agents to securely access and manage backup job details, metrics, client information, and more from Commvault environments. **When to choose:** Choose this if you're using Commvault backup software and want AI agents to manage and monitor your backup infrastructure through natural language commands. **When NOT to choose:** Avoid if you're not using Commvault, as this server is tightly integrated with Commvault APIs and requires a Commvault environment. ### [suse-ai-up](https://mcpcatalogs.com/en/server/suse-suse-ai-up) by SUSE · score 44 · ★28 SUSE AI Universal Proxy is a comprehensive platform for managing and proxying MCP servers with secure orchestration across multiple microservices. **When to choose:** Choose SUSE AI Universal Proxy when you need to manage multiple MCP servers through a unified gateway with authentication, session management, and plugin extensibility in an enterprise Kubernetes environment. **When NOT to choose:** Avoid this solution if you need a lightweight, simple MCP proxy without Kubernetes dependencies or if you're looking for a vendor-agnostic solution not tied to SUSE's ecosystem. ### [mcp](https://mcpcatalogs.com/en/server/yandex-cloud-mcp) by yandex-cloud · score 44 · ★30 A collection of MCP servers for managing Yandex Cloud infrastructure including deployment, serverless, and data platform services. **When to choose:** Choose these servers if you're already using Yandex Cloud and want to integrate its services directly into AI workflows or manage cloud infrastructure through natural language interfaces. **When NOT to choose:** Avoid these servers if you need multi-cloud support or are locked into competing cloud providers, as this solution is vendor-specific to Yandex Cloud with limited interoperability. ### [rootcause](https://mcpcatalogs.com/en/server/yindia-rootcause) by yindia · score 44 · ★31 RootCause is a local-first MCP server that turns natural language requests into Kubernetes incident analysis, diagnostics, and safer operations. **When to choose:** Choose RootCause for Kubernetes operations teams that need evidence-first incident analysis and robust safety controls without compromising on AI-powered diagnostics. **When NOT to choose:** Don't choose RootCause if you're working exclusively with non-Kubernetes environments or need broader cloud provider support beyond AWS integration. ### [mcp-center](https://mcpcatalogs.com/en/server/nautilus-ops-mcp-center) by nautilus-ops · score 42 · ★41 MCP Center is a centralized proxy service for managing and connecting multiple MCP servers with service discovery and authentication. **When to choose:** Choose MCP Center if you're managing multiple MCP servers in a Kubernetes environment and need centralized service discovery, authentication, and lifecycle management. **When NOT to choose:** Avoid MCP Center if you only have one MCP server to manage or if you prefer a simpler solution without the overhead of maintaining a proxy service. ### [toolhive-registry-server](https://mcpcatalogs.com/en/server/stacklok-toolhive-registry-server) by stacklok · score 42 · ★16 A registry server that discovers, governs and controls access to MCP servers and skills across organizations. **When to choose:** Choose this when you need to govern MCP server access in a multi-team environment and require audit trails for compliance. **When NOT to choose:** Not suitable for simple single-user setups as it's designed for enterprise governance with additional complexity. ### [OpsLens](https://mcpcatalogs.com/en/server/sherin-sef-ai-opslens) by Sherin-SEF-AI · score 40 · ★6 OpsLens transforms Notion into an AI-powered incident command center using MCP for automated incident response. **When to choose:** Teams already using Notion for incident management who want AI-powered automation for alert triage, correlation, and response without switching platforms. **When NOT to choose:** Teams not using Notion as their primary workspace or those requiring integration with incident management systems beyond what Notion MCP provides. --- ## other (4) ### [mlb-mcp](https://mcpcatalogs.com/en/server/etweisberg-mlb-mcp) by etweisberg · score 44 · ★25 Python MCP server for MLB baseball analytics with API access to statcast, fangraphs, and baseball reference data. **When to choose:** Choose this server if you need comprehensive baseball statistics access through multiple data sources and want to integrate them into MCP-compatible applications. **When NOT to choose:** Avoid this if you need access to sports data outside of baseball, as it's specialized solely for MLB statistics and won't support other sports analytics. ### [sh-disney-mcp](https://mcpcatalogs.com/en/server/syyuan14-sh-disney-mcp) by syyuan14 · score 43 · ★81 An MCP server that provides real-time ticket pricing and availability for Shanghai Disneyland. **When to choose:** Choose this when you need to integrate Shanghai Disneyland ticket information into an AI application or chatbot for travel planning services. **When NOT to choose:** Don't choose this if you need ticket purchasing functionality, international Disney park data, or non-Shanghai Disneyland China resort information. ### [EVE-Online-OSINT-MCP](https://mcpcatalogs.com/en/server/kongyo2-eve-online-osint-mcp) by kongyo2 · score 38 · ★10 An MCP server providing OSINT tools for EVE Online characters, corporations, and alliances through ESI, EveWho, and zKillboard APIs. **When to choose:** Choose this MCP server if you need comprehensive OSINT capabilities specifically for EVE Online entities, especially for analyzing player behavior, organizational structures, and PvP activities. **When NOT to choose:** Avoid this server if you need OSINT capabilities for games other than EVE Online, or if you require real-time data updates beyond the API providers' refresh rates. ### [EVE-EST-MCP](https://mcpcatalogs.com/en/server/kongyo2-eve-est-mcp) by kongyo2 · score 32 · ★2 EVE-EST-MCP is an MCP server that provides EVE Server Time and downtime calculations for EVE Online players. **When to choose:** Choose this if you need EVE Online server time information and maintenance schedules for automation or integration purposes. **When NOT to choose:** Not suitable if you need general time information or are looking for a server with broader functionality beyond EVE Online-specific data. --- ## productivity (105) ### [time](https://mcpcatalogs.com/en/server/modelcontextprotocol-servers-time) by modelcontextprotocol · score 77 · ★85,748 · official A comprehensive MCP server providing time and timezone conversion tools with automatic system timezone detection. **When to choose:** Choose this server when your LLM application needs to handle time-related queries and timezone conversions, especially for users across different geographic regions. **When NOT to choose:** Avoid if you need more complex time operations beyond basic current time retrieval and timezone conversions, or if you require integration with calendar systems. ### [google_workspace_mcp](https://mcpcatalogs.com/en/server/taylorwilsdon-google-workspace-mcp) by taylorwilsdon · score 58 · ★2,410 Comprehensive Google Workspace MCP server for Gmail, Calendar, Drive, Docs, Sheets, and more with OAuth 2.1 support. **When to choose:** Choose this server when you need comprehensive, fine-grained control over Google Workspace services with secure OAuth 2.1 support for organizational deployment. **When NOT to choose:** Avoid this if you only need access to a few Google services, as it's designed for comprehensive integration which might be overkill for simple use cases. ### [excel-mcp-server](https://mcpcatalogs.com/en/server/haris-musa-excel-mcp-server) by haris-musa · score 58 · ★3,831 A robust MCP server enabling AI agents to manipulate Excel files without Microsoft Excel. **When to choose:** Choose this server when you need AI agents to manipulate Excel files without requiring Excel installations, or when working with Excel-based data processing workflows in automated systems. **When NOT to choose:** Avoid if you need to work with Excel VBA macros, complex Excel add-ins, or if you require very large-scale Excel file processing beyond what Python libraries can handle efficiently. ### [mcp-google-sheets](https://mcpcatalogs.com/en/server/xing5-mcp-google-sheets) by xing5 · score 54 · ★864 A production-ready MCP server for Google Sheets integration with comprehensive tool support and authentication options. **When to choose:** Choose this server when you need comprehensive Google Sheets integration with an MCP client, especially if you're already using Google Workspace and need AI-driven spreadsheet automation. **When NOT to choose:** Avoid this if you need access to non-Google spreadsheet services or if you're concerned about Google API quotas and rate limits for heavy automation workflows. ### [zen](https://mcpcatalogs.com/en/server/sheshbabu-zen) by sheshbabu · score 53 · ★1,085 Selfhosted notes app with MCP integration for searching, listing and reading notes. **When to choose:** Choose Zen if you need a lightweight, self-hosted note-taking solution with markdown support and experimental MCP integration. **When NOT to choose:** Don't choose this if you're looking for a dedicated MCP server implementation rather than a notes application with MCP features. ### [keeper.sh](https://mcpcatalogs.com/en/server/ridafkih-keeper-sh) by ridafkih · score 53 · ★1,067 Keeper.sh is a calendar sync tool with an MCP server for AI agents to access and control multiple calendars. **When to choose:** Choose Keeper when you need to synchronize multiple calendars across different providers and want AI agents to have access to this unified calendar system. **When NOT to choose:** Don't choose Keeper if you need advanced calendar features beyond basic sync, or if you require a self-hosted solution without external dependencies like PostgreSQL and Redis. ### [wenyan-mcp](https://mcpcatalogs.com/en/server/caol64-wenyan-mcp) by caol64 · score 53 · ★1,225 Wenyan MCP Server enables AI to automatically format Markdown and publish to WeChat Official Accounts. **When to choose:** Choose wenyan-mcp if you're already using Claude Desktop and need to publish content directly to WeChat Official Account with AI assistance. **When NOT to choose:** Don't choose this if you need to publish to platforms other than WeChat, or if you can't configure WeChat API credentials. ### [obsidian-mcp-server](https://mcpcatalogs.com/en/server/cyanheads-obsidian-mcp-server) by cyanheads · score 52 · ★525 MCP server for Obsidian vaults with 14 tools for reading, writing, searching, and editing notes, tags, and frontmatter. **When to choose:** Choose this MCP server when you need comprehensive, programmatic access to Obsidian vaults for AI integration or automation, especially when working with complex note structures and metadata. **When NOT to choose:** Avoid if you need read/write access to non-Obsidian note-taking systems or if you require authentication/encryption for sensitive data, as this server has no built-in security model. ### [n8n-workflow-builder](https://mcpcatalogs.com/en/server/makafeli-n8n-workflow-builder) by makafeli · score 52 · ★519 MCP server connecting AI assistants to n8n workflows for natural language automation management. **When to choose:** Teams already using n8n who want to leverage AI assistants for natural language workflow management and automation without UI interaction. **When NOT to choose:** You're not using n8n as your automation platform, or you need workflow management capabilities across multiple different automation tools simultaneously. ### [mcp](https://mcpcatalogs.com/en/server/mondaycom-mcp) by mondaycom · score 52 · ★403 monday.com MCP server provides secure API access to monday.com's work OS for AI agents. **When to choose:** Choose this MCP if you're already using monday.com and want AI agents to directly interact with your work OS, create items, manage boards, and automate workflows through familiar AI interfaces. **When NOT to choose:** Avoid this if you're not a monday.com user or need extensive write access beyond what's provided, as it creates vendor lock-in to the monday.com platform. ### [apple-mcp](https://mcpcatalogs.com/en/server/supermemoryai-apple-mcp) by supermemoryai · score 52 · ★3,090 Apple MCP server that turns native Apple apps into AI superpowers via tools for Messages, Notes, Contacts, Mail, etc. **When to choose:** Choose this if you're a macOS user wanting to automate interactions between your Apple ecosystem applications and AI assistants. **When NOT to choose:** Don't choose this if you're not using macOS, as it requires Apple's native applications to function properly. ### [obsidian-mcp-tools](https://mcpcatalogs.com/en/server/jacksteamdev-obsidian-mcp-tools) by jacksteamdev · score 52 · ★811 An MCP server that bridges Obsidian vaults with AI assistants like Claude for semantic search and template execution. **When to choose:** Only consider if you specifically need this archived version for some compatibility reason, though alternatives are strongly recommended. **When NOT to choose:** Do not use this server as it has been archived and is no longer maintained. There are multiple active alternatives available in the Obsidian Community Plugins store. ### [mcp-mermaid](https://mcpcatalogs.com/en/server/hustcc-mcp-mermaid) by hustcc · score 51 · ★555 An MCP server that generates Mermaid diagrams and charts with AI dynamically. **When to choose:** Choose MCP Mermaid when you need to generate diagrams within AI applications and want multiple output format options, including public sharing via mermaid.ink links. **When NOT to choose:** Don't choose this if you need advanced chart types beyond Mermaid's capabilities or require real-time collaborative editing features. ### [google-workspace-mcp](https://mcpcatalogs.com/en/server/aaronsb-google-workspace-mcp) by aaronsb · score 51 · ★150 A production-ready MCP server providing authenticated access to Google Workspace APIs including Gmail, Calendar, Drive, Sheets, and more. **When to choose:** Choose this MCP server when you need comprehensive, authenticated access to Google Workspace services through AI agents, especially if you're already using Google Workspace APIs and want to leverage the manifest-driven architecture for easy expansion. **When NOT to choose:** Don't choose this if you need access to non-Google services or are concerned about vendor lock-in to Google's ecosystem, as it's specifically designed for Google Workspace APIs. ### [hevy-mcp](https://mcpcatalogs.com/en/server/chrisdoc-hevy-mcp) by chrisdoc · score 51 · ★230 MCP server for Hevy fitness app enabling AI to manage workouts, routines, and exercise templates. **When to choose:** Choose hevy-mcp if you're a Hevy PRO user who wants to manage workouts and routines through AI assistants like Claude Desktop or Cursor. **When NOT to choose:** Avoid if you don't have a Hevy PRO subscription, as the API requires this tier for access. ### [nextcloud-mcp-server](https://mcpcatalogs.com/en/server/cbcoutinho-nextcloud-mcp-server) by cbcoutinho · score 51 · ★223 Production-ready MCP server connecting AI assistants to Nextcloud data with 110+ tools across 10+ apps. **When to choose:** Choose this if you're already using Nextcloud and want AI assistants to interact with your personal or organizational data through a secure, production-ready MCP implementation. **When NOT to choose:** Avoid if you need write access capabilities beyond what Nextcloud provides, or if you prefer not to set up additional infrastructure for optional features like semantic search. ### [applescript-mcp](https://mcpcatalogs.com/en/server/peakmojo-applescript-mcp) by peakmojo · score 51 · ★457 MCP server for AppleScript execution on Mac, enabling AI control of system applications and files. **When to choose:** Choose this when you need AI to control macOS applications and system functions through AppleScript execution, especially for productivity tasks like calendar management and file operations. **When NOT to choose:** Avoid if you need cross-platform support as this only works on macOS, or if you require more system-level access beyond what AppleScript provides. ### [anki-mcp-server](https://mcpcatalogs.com/en/server/nailuogg-anki-mcp-server) by nailuoGG · score 51 · ★239 An MCP server for Anki flashcards via AnkiConnect, allowing LLMs to create, manage, and sync cards. **When to choose:** Choose this server if you want to integrate Anki flashcard management into AI workflows, especially for automating note creation and deck management. **When NOT to choose:** Don't choose this if you're not using Anki or need alternative flashcard systems, as this server is specifically designed to interface with Anki via AnkiConnect. ### [mcp-n8n-workflow-builder](https://mcpcatalogs.com/en/server/salacoste-mcp-n8n-workflow-builder) by salacoste · score 50 · ★226 MCP server enabling AI-powered n8n workflow automation through natural language with 17 tools for comprehensive workflow lifecycle management. **When to choose:** Choose this if you're already using n8n for workflow automation and want to create and manage workflows through conversational AI without leaving your IDE. **When NOT to choose:** Don't choose this if you're not already using n8n, as it's tightly coupled to the n8n ecosystem and requires an existing n8n instance with API access. ### [CoWork-OS](https://mcpcatalogs.com/en/server/cowork-os-cowork-os) by CoWork-OS · score 50 · ★305 CoWork OS is a local-first personal agentic OS that integrates 44 MCP connectors for comprehensive productivity, coding, and knowledge work. **When to choose:** Choose CoWork OS if you're looking for an all-in-one productivity platform that integrates with multiple MCP services and provides a unified workspace for various tasks. **When NOT to choose:** Don't choose CoWork OS if you're specifically looking for an MCP server to extend Claude's capabilities, as it's a client application that consumes MCP services rather than providing them. ### [mcp-outline](https://mcpcatalogs.com/en/server/vortiago-mcp-outline) by Vortiago · score 50 · ★147 MCP server that connects AI assistants to Outline documentation platforms with comprehensive document and collection management tools. **When to choose:** Choose this server if you're already using Outline for documentation and want AI assistants to interact with your documents through MCP. **When NOT to choose:** Don't choose this if you're not using Outline or need document management capabilities for a different platform. ### [raindrop-mcp](https://mcpcatalogs.com/en/server/adeze-raindrop-mcp) by adeze · score 50 · ★163 MCP server for Raindrop.io bookmark management with comprehensive tools for organizing, searching, and editing bookmarks through AI assistants. **When to choose:** Choose this if you use Raindrop.io for bookmark management and want AI assistants to help organize and search your bookmarks through the MCP protocol. **When NOT to choose:** Don't choose this if you need to manage bookmarks outside of Raindrop.io or if you're concerned about vendor lock-in to the Raindrop ecosystem. ### [plane-mcp-server](https://mcpcatalogs.com/en/server/makeplane-plane-mcp-server) by makeplane · score 50 · ★225 Plane's official MCP server provides comprehensive tools for project management, work items, cycles, modules, and more through multiple transport protocols. **When to choose:** Choose this MCP server if you're already using Plane for project management and want AI agents to interact with your Plane workspace programmatically. **When NOT to choose:** Don't choose this if you need write access beyond what Plane's API provides, or if you're concerned about vendor lock-in to the Plane ecosystem. ### [affine-mcp-server](https://mcpcatalogs.com/en/server/dawncr0w-affine-mcp-server) by DAWNCR0W · score 50 · ★166 Comprehensive MCP server for AFFiNE, exposing 84 tools for workspace, document, and database operations over stdio or HTTP. **When to choose:** Choose this server if you use AFFiNE as your knowledge base and want to integrate AI assistants for document retrieval, database interaction, and workflow automation. **When NOT to choose:** Avoid if you only use browser-local AFFiNE workspaces (not server-backed) or need access to AFFiNE features not exposed through the MCP server API. ### [word-mcp-live](https://mcpcatalogs.com/en/server/ykarapazar-word-mcp-live) by ykarapazar · score 49 · ★86 MCP server with 124 tools for Microsoft Word including live editing, tracked changes, and cross-platform functionality. **When to choose:** Choose this when you need AI to interact with Microsoft Word documents, especially for live editing, tracked changes, or when working with existing Word workflows requiring immediate visual feedback. **When NOT to choose:** Don't choose this if you need support for non-Microsoft document formats, if you're on Linux and require live editing, or if you need document processing without having Microsoft Word installed. ### [scrumboy](https://mcpcatalogs.com/en/server/markrai-scrumboy) by markrai · score 49 · ★241 Scrumboy is a self-hosted Kanban project management tool with MCP support for AI agents. **When to choose:** Choose Scrumboy if you need a self-hosted Kanban solution with AI agent integration that prioritizes privacy and customization without vendor lock-in. **When NOT to choose:** Avoid Scrumboy if you need a SaaS solution, extensive third-party integrations beyond its API/webhooks, or require features beyond Kanban methodology. ### [mcp](https://mcpcatalogs.com/en/server/taskade-mcp) by taskade · score 49 · ★148 Taskade MCP server provides 50+ tools for project management, AI agents, and workflows integration with AI assistants. **When to choose:** Choose Taskade MCP if you already use Taskade for project management and want to integrate AI assistants directly with your workspace, or if you need to generate MCP tools from OpenAPI specifications. **When NOT to choose:** Don't choose Taskade MCP if you don't use Taskade, as this requires both a Taskade account and API key to function properly. ### [OmniFocus-MCP](https://mcpcatalogs.com/en/server/themotionmachine-omnifocus-mcp) by themotionmachine · score 49 · ★200 An MCP server for AI assistants to interact with OmniFocus tasks and projects through natural language. **When to choose:** Choose this if you use OmniFocus on macOS and want AI assistants to directly interact with your task management system through natural language. **When NOT to choose:** Don't choose this if you use a different task manager or operating system, as it's specifically designed for OmniFocus on macOS only. ### [joplin-mcp](https://mcpcatalogs.com/en/server/alondmnt-joplin-mcp) by alondmnt · score 49 · ★112 A FastMCP-based server enabling AI assistants to interact with Joplin notes via 25 optimized tools. **When to choose:** Choose this if you use Joplin as your note-taking app and want AI assistants to directly access, search, and manage your notes without manual export/import workflows. **When NOT to choose:** Don't choose this if you're already using a different note-taking application, as this only works with Joplin's specific API and data structure. ### [markmap-mcp-server](https://mcpcatalogs.com/en/server/jinzcdev-markmap-mcp-server) by jinzcdev · score 49 · ★200 An MCP server that converts Markdown to interactive mind maps with export support. **When to choose:** Choose this when you need to visualize hierarchical information from Markdown content and want interactive mind maps with multiple export options. **When NOT to choose:** Don't choose this if you need to edit mind maps directly or require collaboration features with real-time updates. ### [mcp-server-apple-events](https://mcpcatalogs.com/en/server/fradser-mcp-server-apple-events) by FradSer · score 49 · ★112 A production-ready MCP server providing native macOS integration with Apple Reminders and Calendar via EventKit. **When to choose:** Choose this MCP server if you need deep integration with Apple's native Reminders and Calendar applications on macOS, especially for advanced features like location-based reminders and recurring tasks. **When NOT to choose:** Avoid using this server on non-macOS platforms, as it relies on Apple's EventKit framework and native macOS functionality. It's not suitable for cross-platform task management solutions. ### [home-memory](https://mcpcatalogs.com/en/server/impactjo-home-memory) by impactjo · score 48 · ★75 Home Memory is an MCP server that provides structured, persistent knowledge about your home through natural language interactions. **When to choose:** Choose Home Memory if you want an AI-powered home inventory system that works entirely through conversation with multiple AI clients like Claude or OpenAI assistants. **When NOT to choose:** Don't choose Home Memory if you need web-based remote access, multi-user collaboration, or if you're on macOS/Linux without willingness to build from source. ### [yandex-tracker-mcp](https://mcpcatalogs.com/en/server/aikts-yandex-tracker-mcp) by aikts · score 48 · ★73 An MCP server providing full Yandex Tracker API integration with OAuth2 support and caching capabilities. **When to choose:** Choose this server if you're working with Yandex Tracker and want AI assistants to manage issues, queues, and workflows through natural language commands. **When NOT to choose:** Avoid this if you need integration with non-Yandex project management systems or if you're not already using Yandex Tracker in your workflow. ### [getnote-mcp](https://mcpcatalogs.com/en/server/iswalle-getnote-mcp) by iswalle · score 48 · ★61 MCP server for GetNote, a personal note management tool, enabling AI to create, edit, search and organize notes. **When to choose:** Choose this MCP server if you use Get笔记 as your primary note-taking app and want AI assistants to directly interact with your notes and knowledge bases. **When NOT to choose:** Avoid this if you're not a Get笔记 user, as it requires an account with the service and has vendor lock-in to their API. ### [markplane](https://mcpcatalogs.com/en/server/zerowand01-markplane) by zerowand01 · score 48 · ★165 Markplane is an AI-native project management tool that stores tasks in markdown files and exposes functionality through an MCP server. **When to choose:** Choose Markplane if you want AI assistants to directly manage your project tasks within your repository, using version-controlled markdown files instead of external SaaS tools. **When NOT to choose:** Don't choose Markplane if you need a collaborative project management tool with real-time multi-user editing capabilities, or if you prefer traditional GUI-based project management interfaces. ### [mcp-server-siri-shortcuts](https://mcpcatalogs.com/en/server/dvcrn-mcp-server-siri-shortcuts) by dvcrn · score 48 · ★187 MCP server that lets LLMs interact with macOS Siri shortcuts through listing, opening, and executing them. **When to choose:** Choose this when you need to integrate Siri shortcuts into LLM workflows on macOS and want to leverage existing automation. **When NOT to choose:** Don't use this if you're not on macOS or need cross-platform automation capabilities. ### [bear-notes-mcp](https://mcpcatalogs.com/en/server/vasylenko-bear-notes-mcp) by vasylenko · score 48 · ★192 An MCP server for Bear Notes with ranked search, SQLite queries, and write capabilities. **When to choose:** Choose this for Bear Notes users with large libraries who need relevance-ranked search and want to query notes without keeping Bear app open, especially when working across multiple MCP clients. **When NOT to choose:** Avoid if you need write access by default (Edit Mode is off by default) or if you're not on macOS, as Bear is macOS-only. ### [dottedsign-mcp](https://mcpcatalogs.com/en/server/dottedsign-official-dottedsign-mcp) by DottedSign-Official · score 47 · ★116 An MCP server enabling AI assistants to manage DottedSign document signing tasks and templates through natural language. **When to choose:** Choose this if you already use DottedSign for document signing and want to integrate those workflows into your AI assistant. **When NOT to choose:** Don't choose this if you need document signing functionality outside of DottedSign, or if you're looking for open-source/self-hosted solutions. ### [evernote-mcp-server](https://mcpcatalogs.com/en/server/brentmid-evernote-mcp-server) by brentmid · score 47 · ★51 Evernote MCP server for accessing notes via OAuth, with Docker support and recent updates. **When to choose:** Choose this if you need AI-powered natural language search over your personal Evernote notes and want a secure, containerized solution. **When NOT to choose:** Avoid this if you need write access to Evernote notes or prefer a different note-taking platform like Notion. ### [vikunja-mcp](https://mcpcatalogs.com/en/server/democratize-technology-vikunja-mcp) by democratize-technology · score 47 · ★69 MCP server for Vikunja task management with comprehensive CRUD operations, authentication, and architectural improvements. **When to choose:** Choose this MCP server if you're already using Vikunja for task management and want to integrate AI assistants directly into your workflow without leaving the platform. **When NOT to choose:** Don't choose this if you're using a different task management system, as it only works with Vikunja instances. Also consider alternatives if you need more advanced AI capabilities beyond task management. ### [erpnext-mcp-server](https://mcpcatalogs.com/en/server/rakeshgangwar-erpnext-mcp-server) by rakeshgangwar · score 47 · ★90 Connect AI assistants to ERPNext/Frappe systems via MCP for document management and API calls. **When to choose:** Choose this if your team uses ERPNext and wants AI assistants to directly access and manipulate ERP data without custom development. **When NOT to choose:** Don't choose this if you need integration with non-ERPNext systems or if your ERPNext instance doesn't expose API access. ### [activitywatch-mcp-server](https://mcpcatalogs.com/en/server/8bitgentleman-activitywatch-mcp-server) by 8bitgentleman · score 46 · ★65 MCP server for ActivityWatch time tracking data, allowing LLMs to query and interact with time tracking information. **When to choose:** Choose this server if you use ActivityWatch for time tracking and want to query your data through natural language prompts in Claude. **When NOT to choose:** Avoid this if you don't use ActivityWatch, need real-time data access, or require a more comprehensive analytics solution. ### [bear-notes-mcp](https://mcpcatalogs.com/en/server/bejaminjones-bear-notes-mcp) by bejaminjones · score 46 · ★34 MCP server for Bear note app with full read/write capabilities and hybrid sync-safe approach. **When to choose:** Choose this if you're a Bear notes user who wants comprehensive read/write access to your notes through Claude Desktop with iCloud sync safety. **When NOT to choose:** Don't choose this if you don't use the Bear app, as it's specifically designed for Bear's database structure and API. ### [nutrient-document-engine-mcp-server](https://mcpcatalogs.com/en/server/pspdfkit-nutrient-document-engine-mcp-server) by PSPDFKit · score 46 · ★59 MCP server for document processing with text extraction, redaction, and editing capabilities through natural language. **When to choose:** Teams needing to process documents through natural language interactions with AI, particularly those already using PDFs and forms requiring extraction, editing, or redaction. **When NOT to choose:** You shouldn't choose this if you need an open-source solution without vendor lock-in, as it requires purchasing a license for production use beyond evaluation mode. ### [OpenStudy](https://mcpcatalogs.com/en/server/openstudy-dev-openstudy) by OpenStudy-dev · score 46 · ★66 A self-hostable study dashboard with a built-in MCP server that enhances Claude's effectiveness for academic tasks. **When to choose:** Choose this if you're a student who wants to integrate AI assistants directly with your study workflow and prefer self-hosted solutions. **When NOT to choose:** Don't choose this if you need multi-user collaboration features or prefer cloud-based study management without self-hosting requirements. ### [mcp-bear](https://mcpcatalogs.com/en/server/jkawamoto-mcp-bear) by jkawamoto · score 46 · ★70 MCP server for Bear note-taking app with multiple client support and good documentation. **When to choose:** Choose this MCP server if you use Bear for note-taking and want AI assistants to interact with your notes through the MCP protocol. **When NOT to choose:** Don't choose this if you use a different note-taking app or if you need full write capabilities to notes (some actions are partial). ### [wanderlog-mcp](https://mcpcatalogs.com/en/server/shaikhspeare-wanderlog-mcp) by shaikhspeare · score 46 · ★27 MCP server for building and editing trip itineraries in Wanderlog through natural language conversation. **When to choose:** 选择此服务如果你已经是 Wanderlog 用户并希望通过自然语言对话而非手动点击界面来规划和管理旅行行程。 **When NOT to choose:** 如果你不使用 Wanderlog 平台，或者需要更广泛的旅行规划工具集成，这个服务器可能不适合你。 ### [lapras-mcp-server](https://mcpcatalogs.com/en/server/lapras-inc-lapras-mcp-server) by lapras-inc · score 46 · ★101 An official MCP server for Lapras, enabling job search and career management through AI. **When to choose:** Choose this server if you're actively job hunting or managing your career on lapras.com and want AI assistance with job search and profile management. **When NOT to choose:** Don't choose this if you're not using lapras.com for job hunting, as it requires their platform to be useful. ### [better-bear](https://mcpcatalogs.com/en/server/kuvopllc-better-bear) by KuvopLLC · score 46 · ★58 An MCP server for Bear notes with CLI tool, cloud sync, and context library. **When to choose:** Choose better-bear if you're a Bear notes user who wants seamless integration with Claude and need an efficient context library pattern for LLM consumption. **When NOT to choose:** Don't choose better-bear if you don't use Bear notes, are concerned about vendor lock-in to Apple's ecosystem, or need cross-platform note app support. ### [caldav-mcp](https://mcpcatalogs.com/en/server/dominik1001-caldav-mcp) by dominik1001 · score 46 · ★69 A CalDAV MCP server enabling AI assistants to manage calendar events via CalDAV protocols. **When to choose:** Choose this when you need an AI assistant to interact with your existing CalDAV calendar service without additional infrastructure. **When NOT to choose:** Don't choose this if you need OAuth authentication, integration with non-CalDAV calendar services, or if you're already using a different calendar MCP server. ### [FocusRelayMCP](https://mcpcatalogs.com/en/server/deverman-focusrelaymcp) by deverman · score 46 · ★25 MCP server for OmniFocus that allows querying tasks, projects, and tags using natural language. **When to choose:** Choose FocusRelayMCP if you're an OmniFocus user who wants natural language access to your task management system without leaving your AI assistant interface. **When NOT to choose:** Avoid if you need write access to OmniFocus (this is read-only), use a different task management system, or don't want to install an additional plugin in OmniFocus. ### [Zammad-MCP](https://mcpcatalogs.com/en/server/basher83-zammad-mcp) by basher83 · score 46 · ★30 A comprehensive MCP server for Zammad integration with extensive ticket management tools, attachment support, and both stdio/HTTP transport options. **When to choose:** Choose this if you're using Zammad as your helpdesk system and want AI assistants to directly interact with tickets, users, and attachments. **When NOT to choose:** Don't choose this if you're not using Zammad, as there's no benefit without an existing Zammad instance to integrate with. ### [mcp-server](https://mcpcatalogs.com/en/server/aayush9029-mcp-server) by Aayush9029 · score 45 · ★42 A task management MCP server that connects AI assistants to iPhone task tracking with notifications. **When to choose:** Choose this server when you're using AI assistants for development and want seamless integration with iPhone-based task management and notifications. **When NOT to choose:** Avoid this if you need cross-platform task management beyond iOS or prefer open-source self-hosted solutions without vendor dependencies. ### [dav-mcp](https://mcpcatalogs.com/en/server/philflowio-dav-mcp) by PhilflowIO · score 45 · ★19 An MCP server providing 26 tools for calendar, contact, and task management via CalDAV, CardDAV, and VTODO protocols. **When to choose:** Choose dav-mcp when you need comprehensive calendar, contact, and task management capabilities for AI agents with full CRUD operations and server-side filtering across multiple protocols. **When NOT to choose:** Avoid if you only need basic calendar operations or are looking for a solution with a managed cloud backend rather than self-hosted servers. ### [mcp-toggl](https://mcpcatalogs.com/en/server/verygoodplugins-mcp-toggl) by verygoodplugins · score 45 · ★17 MCP server for Toggl Track integration with intelligent caching, time tracking, and activity reporting tools. **When to choose:** Choose this server if you use Toggl Track for time tracking and want to integrate AI-powered insights directly into your workflow without manual report generation. **When NOT to choose:** Avoid this if you need time tracking solutions outside Toggl, require write access to time tracking data, or have privacy concerns about sharing desktop activity with AI systems. ### [miro-mcp-server](https://mcpcatalogs.com/en/server/olgasafonova-miro-mcp-server) by olgasafonova · score 45 · ★22 Comprehensive MCP server for controlling Miro whiteboards with 92 tools, featuring token efficiency options and workflow skills. **When to choose:** Choose this server when you need comprehensive programmatic control over Miro whiteboards and want to integrate whiteboard functionality into AI workflows. **When NOT to choose:** Don't choose this if you need read-only access to Miro boards, require features not available in Miro's API, or need more than the 92 supported tools. ### [paperless-mcp](https://mcpcatalogs.com/en/server/nloui-paperless-mcp) by nloui · score 45 · ★173 An MCP server that provides comprehensive document management capabilities for Paperless-NGX document management system. **When to choose:** Choose this if you're using Paperless-NGX for document management and want to leverage AI for organizing, tagging, and manipulating documents through natural language. **When NOT to choose:** Don't choose this if you're using a different document management system, as this MCP server is specifically designed for Paperless-NGX integration only. ### [clickup-mcp-server](https://mcpcatalogs.com/en/server/taazkareem-clickup-mcp-server) by taazkareem · score 45 · ★39 A premium MCP server for integrating ClickUp project management with AI agents through task management, document handling, and time tracking. **When to choose:** Choose this if you're already using ClickUp for project management and want deep AI integration with its features like tasks, time tracking, and documents. **When NOT to choose:** Don't choose this if you need open-source alternatives, as this project has transitioned to a paid model with licensing requirements. ### [mcp-rss-aggregator](https://mcpcatalogs.com/en/server/imprvhub-mcp-rss-aggregator) by imprvhub · score 44 · ★25 An MCP server that fetches and reads RSS feeds in Claude Desktop with OPML support and article filtering. **When to choose:** Choose this if you want to consume news and content from multiple RSS feeds directly within Claude Desktop without switching applications. **When NOT to choose:** Don't choose this if you need RSS filtering features beyond categories or if you prefer a full-featured RSS reader with advanced content organization. ### [streamdeck-mcp](https://mcpcatalogs.com/en/server/verygoodplugins-streamdeck-mcp) by verygoodplugins · score 44 · ★13 MCP server for Elgato Stream Deck control, letting AI create custom profiles with buttons, icons, and scripts. **When to choose:** Choose this if you want to automate Stream Deck configuration through AI and need integration with other MCP services like Slack or Home Assistant. **When NOT to choose:** Don't choose this if you need direct hardware control without the Elgato desktop app, or if you're on Linux and prefer the legacy USB mode. ### [mcp-ical](https://mcpcatalogs.com/en/server/omar-v2-mcp-ical) by Omar-V2 · score 44 · ★311 MCP server for natural language macOS calendar management with event creation and scheduling. **When to choose:** Choose this if you're a macOS user who wants natural language calendar management and are already using Claude or another MCP-compatible client. **When NOT to choose:** Don't choose this if you're not using macOS, as it's specifically designed for the macOS Calendar app and has no Windows or Linux support. ### [silverbullet-mcp](https://mcpcatalogs.com/en/server/ahmad-a0-silverbullet-mcp) by Ahmad-A0 · score 44 · ★31 A TypeScript MCP server enabling LLMs to interact with SilverBullet notes via tools and resources. **When to choose:** Choose this MCP server if you use SilverBullet as your knowledge base and want AI assistants to access and query your notes. **When NOT to choose:** Don't choose this if you don't use SilverBullet or prefer a different note-taking system with MCP integration. ### [mcp-simple-timeserver](https://mcpcatalogs.com/en/server/andybrandt-mcp-simple-timeserver) by andybrandt · score 44 · ★30 A Python MCP server providing time queries, holiday information, and time calculations for Claude. **When to choose:** Choose this server when your application needs accurate time data, date calculations, and holiday information across multiple countries and calendar systems. **When NOT to choose:** Avoid if you need highly customizable time formatting beyond what's provided, or if you need historical time data before 1970 (Unix epoch limitation). ### [che-ical-mcp](https://mcpcatalogs.com/en/server/psychquant-che-ical-mcp) by PsychQuant · score 44 · ★27 macOS Calendar & Reminders MCP server with 29 tools and native Swift EventKit integration. **When to choose:** Choose this server for macOS users who need deep Calendar and Reminders integration with advanced features like tagging, batch operations, conflict detection, and undo/redo functionality. **When NOT to choose:** Not suitable for non-macOS users, or those who need cross-platform calendar solutions with vendor-agnostic approaches. ### [autotask-mcp](https://mcpcatalogs.com/en/server/wyre-technology-autotask-mcp) by wyre-technology · score 44 · ★36 MCP server providing 39 tools for Kaseya Autotask PSA integration, enabling AI to manage tickets, projects, time entries, and companies through natural language. **When to choose:** Choose this if you're an MSP using Autotask PSA and want to eliminate context switching between Autotask and your AI assistant. **When NOT to choose:** Avoid if you need access to Autotask features not covered by these 39 tools, or if you require advanced UI interactions beyond API access. ### [Deskdrop](https://mcpcatalogs.com/en/server/svreenen-deskdrop) by SvReenen · score 44 · ★27 Android keyboard with MCP support for smart home control via Home Assistant **When to choose:** Choose Deskdrop if you want an AI-enhanced keyboard that integrates with your smart home system through MCP and prioritizes privacy with local processing options. **When NOT to choose:** Don't choose Deskdrop if you need keyboard functionality on iOS, require commercial support, or prefer solutions available directly through app stores. ### [taskflow-mcp](https://mcpcatalogs.com/en/server/pinkpixel-dev-taskflow-mcp) by pinkpixel-dev · score 44 · ★29 TaskFlow MCP is a specialized server that helps AI assistants break down user requests into manageable tasks with subtasks, dependencies, and notes. **When to choose:** Choose this server when you need structured task management with subtasks, dependencies, and user approval workflows for complex projects. **When NOT to choose:** Avoid this server if you need a simple to-do list without subtasks or if you require real-time collaboration features. ### [apple-notes-mcp](https://mcpcatalogs.com/en/server/sweetrb-apple-notes-mcp) by sweetrb · score 44 · ★27 MCP server that bridges AI assistants with Apple Notes on macOS via AppleScript. **When to choose:** Choose this if you use Apple Notes on macOS and want AI assistance to create, search, and organize your notes with full privacy since all processing stays local. **When NOT to choose:** Don't choose this if you need cross-platform support or use a note-taking app other than Apple Notes, as this is macOS-specific and requires Apple Notes. ### [mcp-stytch-consumer-todo-list](https://mcpcatalogs.com/en/server/stytchauth-mcp-stytch-consumer-todo-list) by stytchauth · score 44 · ★27 Cloudflare Workers MCP server for TODO list management with Stytch authentication. **When to choose:** Choose this if you need an MCP server with proper authentication for accessing TODO functionality, especially if you're already using or planning to use Stytch for authentication. **When NOT to choose:** Avoid if you need self-hosted solution or don't want vendor lock-in to both Cloudflare Workers and Stytch authentication services. ### [intervals-icu-mcp](https://mcpcatalogs.com/en/server/eddmann-intervals-icu-mcp) by eddmann · score 43 · ★25 Comprehensive MCP server for Intervals.icu integration with 48 tools for training data, wellness metrics, and performance analysis. **When to choose:** Choose this server if you're a serious cyclist or runner using Intervals.icu for training data and want AI assistance with training analysis and planning. **When NOT to choose:** Don't choose this if you don't use Intervals.icu for training or prefer a more general-purpose fitness tracking platform integration. ### [clay-mcp](https://mcpcatalogs.com/en/server/clay-inc-clay-mcp) by clay-inc · score 43 · ★31 Clay MCP server enables contact management, CRM operations, and interaction tracking through MCP tools. **When to choose:** Choose this server if you're already using Clay for contact management and want to integrate your contact data with AI assistants for querying, updating, and analyzing your professional network. **When NOT to choose:** Don't choose this server if you need contact management outside of Clay ecosystem, as it requires a Clay account and doesn't integrate with other CRM systems. ### [mcp-tasks](https://mcpcatalogs.com/en/server/flesler-mcp-tasks) by flesler · score 43 · ★45 A comprehensive MCP server for task management supporting Markdown, JSON and YAML formats with AI-optimized features. **When to choose:** Choose this server when you need a budget-optimized task management system specifically designed for AI integration, with robust safety features to prevent task loss. **When NOT to choose:** Don't choose this if you need extensive task manipulation beyond what the 5 core tools provide, or if you require complex project management features like dependencies or time tracking. ### [gsd-task-manager](https://mcpcatalogs.com/en/server/vscarpenter-gsd-task-manager) by vscarpenter · score 43 · ★18 A privacy-first Eisenhower Matrix task manager with MCP server integration for AI-powered task management. **When to choose:** Choose GSD if you're looking for a privacy-first task manager that implements the Eisenhower Matrix methodology and want AI assistance through MCP integration. **When NOT to choose:** Don't choose GSD if you need advanced team collaboration features, as it's primarily designed for individual productivity with limited multi-user support. ### [SupaThings-MCP](https://mcpcatalogs.com/en/server/soycanopa-supathings-mcp) by soycanopa · score 43 · ★21 Things 3 MCP server for macOS that combines SQLite reads with URL writes for AI agents. **When to choose:** Choose SupaThings when you're a macOS user of Things 3 who wants AI agents to understand and manage your task structure beyond simple URL-based commands. **When NOT to choose:** Don't choose this if you're not on macOS, don't use Things 3, or need write capabilities beyond the official Things URL scheme. ### [raindrop-io-mcp-server](https://mcpcatalogs.com/en/server/hiromitsusasaki-raindrop-io-mcp-server) by hiromitsusasaki · score 43 · ★72 MCP server enabling LLMs to create, search, and filter Raindrop.io bookmarks. **When to choose:** Choose this if you use Raindrop.io for bookmark management and want AI assistants to interact with your collection. **When NOT to choose:** Don't choose this if you need access to advanced Raindrop.io features beyond basic bookmark operations, or if you prefer a different bookmark manager. ### [claude-code-google-workspace](https://mcpcatalogs.com/en/server/evolsb-claude-code-google-workspace) by evolsb · score 43 · ★20 MCP servers for Google Workspace (Gmail, Drive, Calendar, Sheets, Docs) and Slack with multi-account support. **When to choose:** Choose this if you need to access multiple Google Workspace accounts and Slack through Claude Code, especially when official integrations aren't available yet. **When NOT to choose:** Don't choose this if you're already using Claude Desktop/Cowork which has built-in Google Workspace integrations, or if you need write access beyond what Google's official CLI provides. ### [mendeley-mcp](https://mcpcatalogs.com/en/server/pallaprolus-mendeley-mcp) by pallaprolus · score 42 · ★19 A functional MCP server connecting Mendeley reference library to Claude Desktop and other MCP clients with search, retrieval, and management capabilities. **When to choose:** Choose this if you're an academic researcher or student who uses Mendeley to manage your literature library and want AI assistance to search and retrieve papers. **When NOT to choose:** Avoid if you don't use Mendeley or prefer different reference management tools like Zotero or EndNote. ### [obsidian-cli-rest](https://mcpcatalogs.com/en/server/dsebastien-obsidian-cli-rest) by dsebastien · score 42 · ★14 Turns Obsidian CLI commands into an HTTP API and MCP server for automation and AI assistant integration. **When to choose:** Choose this if you need to programmatically control your Obsidian vault or connect AI assistants to your notes via MCP. **When NOT to choose:** Don't choose this if you need access to Obsidian GUI features or if you're concerned about potential security risks of exposing your vault via API. ### [miniflux-mcp](https://mcpcatalogs.com/en/server/tssujt-miniflux-mcp) by tssujt · score 42 · ★12 A Go-based MCP server providing 40+ tools for managing Miniflux RSS reader feeds, entries, and categories. **When to choose:** Choose this if you use Miniflux as your RSS reader and want AI assistants to manage your feeds and content through natural language. **When NOT to choose:** Don't choose this if you don't already use Miniflux or prefer other RSS readers like Inoreader or Feedly. ### [zotero-mcp](https://mcpcatalogs.com/en/server/kujenga-zotero-mcp) by kujenga · score 42 · ★150 Python MCP server for Zotero library integration with AI assistants **When to choose:** Choose this if you need to connect AI assistants to your existing Zotero library for research or academic work. **When NOT to choose:** Don't choose this if you need write access to your library (this is read-only) or if you're concerned about relying on Zotero's API stability. ### [Excel-MCP-Server-Master](https://mcpcatalogs.com/en/server/guillehr2-excel-mcp-server-master) by guillehr2 · score 41 · ★28 A comprehensive MCP server for manipulating Excel files without Microsoft Excel, with full CRUD operations, chart creation, and Claude AI integration. **When to choose:** Choose this MCP server when you need to manipulate Excel files without Microsoft Excel installed, particularly for automated report generation through AI assistants. **When NOT to choose:** Don't choose this if you need advanced Excel VBA macros or functions not implemented in the server, or if you require real-time collaboration features of Microsoft 365. ### [Amazing-Marvin-MCP](https://mcpcatalogs.com/en/server/bgheneti-amazing-marvin-mcp) by bgheneti · score 41 · ★34 MCP server connecting Amazing Marvin productivity app with AI assistants for task and project management. **When to choose:** Choose this if you're already using Amazing Marvin for productivity and want to leverage AI assistants for smarter task management and planning without copying data manually. **When NOT to choose:** Don't choose this if you don't use Amazing Marvin, as it requires an active account and API key from the service to function. ### [workflows-mcp-server](https://mcpcatalogs.com/en/server/cyanheads-workflows-mcp-server) by cyanheads · score 41 · ★31 MCP server that enables AI agents to discover, create, and execute complex workflows defined in YAML files. **When to choose:** When you need to structure complex, multi-step AI workflows and want reusable components that can be shared across agents. **When NOT to choose:** If you need real-time workflow execution without YAML overhead or are already committed to a different workflow orchestration system. ### [sn-mcp-server](https://mcpcatalogs.com/en/server/signnow-sn-mcp-server) by signnow · score 40 · ★6 SignNow's official MCP server providing secure e-signature workflows, templates, and document management for AI agents. **When to choose:** Choose this MCP server when your AI agent needs to interact with electronic signature workflows and you're already using or evaluating SignNow as your e-signature provider. **When NOT to choose:** Avoid if you need to integrate with non-SignNow e-signature providers or if you prefer a more general-purpose document automation solution that isn't tied to a specific vendor. ### [autowpmcp](https://mcpcatalogs.com/en/server/njengah-autowpmcp) by Njengah · score 40 · ★21 AutoWP MCP server connects Claude to WordPress sites for automated content creation and site management. **When to choose:** Choose AutoWP MCP if you frequently work with WordPress sites and want to streamline content creation and management directly from AI conversations without switching between interfaces. **When NOT to choose:** Don't choose this if you need to manage non-WordPress sites, require advanced plugin/theme management features, or need granular control over WordPress core updates and site health monitoring. ### [joplin-mcp-server](https://mcpcatalogs.com/en/server/dweigend-joplin-mcp-server) by dweigend · score 40 · ★73 MCP server for Joplin notes with CRUD operations and search functionality. **When to choose:** Choose this if you're already using Joplin as your note-taking system and want to integrate it with AI assistants like Claude for enhanced note interaction and automation. **When NOT to choose:** Avoid this if you need mobile access to your notes, use only cloud-based note-taking solutions, or require features not available in Joplin's Web Clipper Service. ### [mcp-server-memos-py](https://mcpcatalogs.com/en/server/ryojerryyu-mcp-server-memos-py) by RyoJerryYu · score 40 · ★21 Python MCP server enabling LLM interaction with Memos for memo search, creation, and management. **When to choose:** Choose this MCP server if you use Memos for note-taking and want to enable AI agents to interact with your memo collection, search existing notes, or create new ones programmatically. **When NOT to choose:** Avoid this if you're not using Memos, need more complex memo management features, or require support for other note-taking applications like Notion or Evernote. ### [mcp-task-manager-server](https://mcpcatalogs.com/en/server/bsmi021-mcp-task-manager-server) by bsmi021 · score 40 · ★23 An MCP server for project and task management using SQLite with client-driven tools. **When to choose:** Choose this server for local, client-driven task management with SQLite persistence, especially when you need granular control over task relationships and dependencies. **When NOT to choose:** Don't choose this if you need multi-user authentication, cloud-based task management, or integration with external services beyond basic import/export. ### [toolbase](https://mcpcatalogs.com/en/server/toolbase-ai-toolbase) by Toolbase-AI · score 39 · ★167 Toolbase is a desktop application that extends Claude and AI platforms with a visual interface for discovering, installing, and managing MCP-powered tools. **When to choose:** Choose Toolbase if you're a Claude Desktop user who wants a visual interface for discovering and managing MCP tools without technical expertise. **When NOT to choose:** Don't choose Toolbase if you need Windows/Linux support (currently macOS only), or if you prefer configuring MCP servers directly rather than through a graphical interface. ### [TextEdit-MCP](https://mcpcatalogs.com/en/server/m-pineapple-textedit-mcp) by M-Pineapple · score 39 · ★3 A Swift-based MCP server for creating RTF documents with native macOS TextEdit formatting features. **When to choose:** Choose TextEdit-MCP when you need to create richly formatted RTF documents on macOS with native TextEdit compatibility, especially for business or technical documentation. **When NOT to choose:** Don't choose this if you need cross-platform support or require document formats other than RTF, as it's macOS-only and specialized for RTF generation. ### [excalidraw-mcp](https://mcpcatalogs.com/en/server/cmd8-excalidraw-mcp) by cmd8 · score 38 · ★4 MCP server for creating and editing Excalidraw diagrams programmatically with node and edge manipulation tools. **When to choose:** Choose this MCP server when you need AI systems to interact with Excalidraw diagrams programmatically, especially if you already use Excalidraw for your visual documentation or design work. **When NOT to choose:** Don't choose this server if you need to work with diagram formats other than Excalidraw, or if you require advanced diagram manipulation features not provided by this implementation. ### [visa-mcp](https://mcpcatalogs.com/en/server/byigitt-visa-mcp) by byigitt · score 38 · ★9 MCP server for querying Schengen visa appointment status through multiple filter options. **When to choose:** Choose this server when you need quick access to Schengen visa appointment status and have a specific need to filter visas by country, status, mission, or application center. **When NOT to choose:** Don't choose this server if you need visa appointment management functionality beyond status checking, or if the visasbot.com API doesn't cover the specific visa types you need. ### [cronmcp](https://mcpcatalogs.com/en/server/vaibhavpandeyvpz-cronmcp) by vaibhavpandeyvpz · score 38 · ★2 Open-source MCP server that exposes cron scheduling capabilities via the Model Context Protocol. **When to choose:** Choose cronmcp when you need simple, time-based task scheduling within your MCP workflow, especially if you're already using Node.js and need lightweight dependency management. **When NOT to choose:** Avoid cronmcp if you need complex scheduling with timezones, persistent history of job executions, or need to handle more than 50 simultaneous jobs. ### [job-searchoor](https://mcpcatalogs.com/en/server/0xdaef0f-job-searchoor) by 0xDAEF0F · score 37 · ★63 A job search MCP server that provides tools to filter and find relevant job opportunities based on keywords, location, and recency. **When to choose:** Choose Job Searchoor if you're a job seeker who wants to integrate job search directly into your AI workflow without switching between multiple platforms. **When NOT to choose:** Don't choose Job Searchoor if you need specialized job search features like company research, salary comparison, or application tracking that aren't provided. ### [cezzis-com-cocktails-mcp](https://mcpcatalogs.com/en/server/mtnvencenzo-cezzis-com-cocktails-mcp) by mtnvencenzo · score 36 · ★0 Go-based MCP server for Cezzis.com cocktails that exposes tools for searching cocktails and managing user authentication. **When to choose:** Choose this server when you need to integrate cocktail discovery and rating capabilities into your MCP client, especially if you're already using or planning to use the Cezzis platform. **When NOT to choose:** Don't choose this server if you need a self-contained solution without external dependencies, if you're not using the Cezzis platform, or if you require an open-source license (this is proprietary software). ### [unanet-mcp-server](https://mcpcatalogs.com/en/server/culstrup-unanet-mcp-server) by culstrup · score 36 · ★3 MCP server connecting Claude Desktop to Unanet GovCon ERP for natural language project, timesheet, and report management. **When to choose:** Government contractors already using Unanet who want to save time on reporting, status updates, and data retrieval through natural language interactions. **When NOT to choose:** Not for organizations using other ERP systems as this server is specifically designed for Unanet integration only. ### [mcp-kibela-server](https://mcpcatalogs.com/en/server/kiwamizamurai-mcp-kibela-server) by kiwamizamurai · score 36 · ★7 MCP server for Kibela API integration, enabling LLMs to search, retrieve, and manage notes with advanced filtering capabilities. **When to choose:** Choose this server if you use Kibela as your team knowledge base and want AI assistants to be able to search, retrieve, and interact with your documentation and notes. **When NOT to choose:** Don't choose this if you don't use Kibela or need write access beyond liking/unliking notes (most operations are read-only). ### [systemprompt-mcp-taskchecker](https://mcpcatalogs.com/en/server/ejb503-systemprompt-mcp-taskchecker) by Ejb503 · score 35 · ★2 MCP server for intelligent task management, evaluation scoring, and session-based workflow tracking with comprehensive enterprise features. **When to choose:** Choose this server if you need enterprise-grade task management with evaluation scoring and session-based workflows for AI assistants. **When NOT to choose:** Don't choose this if you need persistent task storage beyond session duration or require evaluation metrics beyond the 0-100 scale. ### [bear-mcp](https://mcpcatalogs.com/en/server/autonomee-bear-mcp) by autonomee · score 34 · ★0 Bear Notes MCP server enables AI to search, read, create, and browse Bear notes directly through their local SQLite database. **When to choose:** Choose Bear Notes MCP when you use Bear for note-taking and want AI assistants to access your notes without cloud dependencies. **When NOT to choose:** Don't choose this if you use a different note-taking app, need cross-platform support (Bear is macOS/iOS only), or require advanced note editing capabilities. ### [uberall-mcp-server](https://mcpcatalogs.com/en/server/uberall-uberall-mcp-server) by uberall · score 34 · ★1 Uberall MCP server provides business listing and social media management tools through AI assistants. **When to choose:** Choose this MCP server if you're already using Uberall for business listing management and want to integrate it with your AI workflows for social media posting and location management. **When NOT to choose:** Don't choose this if you need multi-platform business management beyond Uberall, or if you don't have access to an Uberall API account. ### [holaspirit-mcp-server](https://mcpcatalogs.com/en/server/syucream-holaspirit-mcp-server) by syucream · score 33 · ★1 An MCP server providing access to Holaspirit API for organizational governance data. **When to choose:** Choose this server if you're already using Holasprit for organizational management and want AI assistants to access your organizational data without additional integration work. **When NOT to choose:** Don't choose this if you need write access to Holasprit data (this server appears read-only only) or if you're not already committed to the Holasprit platform. ### [google-docs-mcp](https://mcpcatalogs.com/en/server/jag-k-google-docs-mcp) by jag-k · score 33 · ★0 An MCP server for Google Docs that enables reading documents, managing comments, and text manipulation via OAuth authentication. **When to choose:** Choose this when you need AI assistants to interact with Google Docs documents, especially if you're already using Google Workspace and want to maintain your existing document ecosystem. **When NOT to choose:** Don't choose if you need extensive write capabilities beyond simple text insertion, or if you're locked into a different document platform like Notion or Microsoft Word. ### [todo-mcp-server](https://mcpcatalogs.com/en/server/idsulik-todo-mcp-server) by idsulik · score 32 · ★7 A minimal Todo MCP server providing basic task management operations through Model Context Protocol. **When to choose:** Choose this for personal task management within AI assistants or as a reference implementation for learning MCP server development. **When NOT to choose:** Don't choose this for team collaboration needs as it appears to be designed for single-user scenarios with no apparent multi-user support. ### [mcp-google-tasks](https://mcpcatalogs.com/en/server/ktmage-mcp-google-tasks) by ktmage · score 29 · ★2 MCP server for Google Tasks integration, allowing AI assistants to manage tasks through the protocol. **When to choose:** Choose this if you're already using Google Tasks and want AI assistants to manage your tasks through MCP without switching platforms. **When NOT to choose:** Don't choose this if you need Google Calendar integration (this only handles Tasks), or if you prefer a task management platform outside the Google ecosystem. ### [sticky-notes-server](https://mcpcatalogs.com/en/server/bsmi021-sticky-notes-server) by bsmi021 · score 28 · ★1 MCP server for managing sticky notes with REST API, WebSocket support, and React UI. **When to choose:** Choose this server when you need a sticky notes solution that integrates with MCP and offers both a web UI and programmatic access through REST API. **When NOT to choose:** Avoid if you need cloud-based sync across devices or require advanced collaboration features beyond basic real-time synchronization. --- ## search (37) ### [exa-mcp-server](https://mcpcatalogs.com/en/server/exa-labs-exa-mcp-server) by exa-labs · score 59 · ★4,436 MCP server for Exa web search and web crawling capabilities with support for multiple AI clients. **When to choose:** Choose this MCP server when you need reliable web search capabilities for your AI applications, especially if you're already using Exa's services or need well-structured search results with minimal setup. **When NOT to choose:** Don't choose this if you need offline search capabilities, have concerns about vendor lock-in with Exa, or require search functionality that's not covered by Exa's API offerings. ### [ddgs](https://mcpcatalogs.com/en/server/deedy5-ddgs) by deedy5 · score 55 · ★2,640 An MCP server providing web search capabilities across multiple search engines via text, images, videos, news, and books searches. **When to choose:** Choose DDGS when you need a unified interface to multiple search engines through MCP, particularly for AI applications that need comprehensive web search capabilities. **When NOT to choose:** Don't choose DDGS if you need specialized search features not provided by the supported engines or require authenticated access to private search APIs. ### [mcp-searxng](https://mcpcatalogs.com/en/server/ihor-sokoliuk-mcp-searxng) by ihor-sokoliuk · score 54 · ★786 MCP server providing web search capabilities through SearXNG API with intelligent caching and content extraction. **When to choose:** Choose this MCP server when you need privacy-focused web search capabilities with content extraction and intelligent caching for AI assistants. **When NOT to choose:** Don't choose this if you need search results from specific proprietary search engines or require direct integration without a SearXNG instance. ### [hister](https://mcpcatalogs.com/en/server/asciimoo-hister) by asciimoo · score 52 · ★959 Hister is a privacy-focused search engine with MCP support for AI-enhanced querying. **When to choose:** Choose Hister when you need a privacy-focused search solution that can be integrated with AI agents via MCP for enhanced search capabilities. **When NOT to choose:** Don't choose Hister if you need enterprise-level features, user authentication, or require cloud-hosted search with high scalability. ### [annas-mcp](https://mcpcatalogs.com/en/server/iosifache-annas-mcp) by iosifache · score 52 · ★863 An MCP server and CLI tool for searching and downloading documents from Anna's Archive. **When to choose:** Choose this server when you need to programmatically search and retrieve documents from Anna's Archive through an MCP interface, particularly for research or academic applications. **When NOT to choose:** Don't choose this if you need access to copyrighted materials without proper authorization, or if you're looking for a general-purpose document search solution beyond Anna's Archive's scope. ### [mcp-omnisearch](https://mcpcatalogs.com/en/server/spences10-mcp-omnisearch) by spences10 · score 51 · ★308 A MCP server providing unified access to multiple search engines and AI tools through a single interface. **When to choose:** Choose this server when you need unified access to multiple search engines and AI tools without switching between different interfaces. **When NOT to choose:** Don't choose this if you need specialized search functionality not supported by the integrated providers, or if you're concerned about vendor lock-in to multiple services. ### [meilisearch-mcp](https://mcpcatalogs.com/en/server/meilisearch-meilisearch-mcp) by meilisearch · score 50 · ★186 Meilisearch MCP server enables LLMs to interact with Meilisearch search engine through natural language commands. **When to choose:** Choose this when you need an AI assistant to manage and query Meilisearch search indices without requiring developers to learn the search API directly. **When NOT to choose:** Don't choose this if you need advanced security controls for API management, as the README notes the chat-based key addition is primarily for development purposes. ### [google-surf-mcp](https://mcpcatalogs.com/en/server/harimxchoi-google-surf-mcp) by HarimxChoi · score 50 · ★169 A robust Google search MCP server combining search, content extraction, and academic paper analysis without API keys. **When to choose:** Choose this when you need a complete search, fetching, and extraction solution without API keys and want CAPTCHA handling built-in. **When NOT to choose:** Don't choose for high-volume applications or serverless environments without manual CAPTCHA solving capability. ### [web-scout-mcp](https://mcpcatalogs.com/en/server/pinkpixel-dev-web-scout-mcp) by pinkpixel-dev · score 49 · ★128 A robust MCP server providing web search and content extraction capabilities through DuckDuckGo integration. **When to choose:** Choose Web Scout when you need a privacy-focused search solution with efficient content extraction capabilities for your MCP-enabled AI assistant. **When NOT to choose:** Avoid if you need search results from Google or require advanced content filtering beyond basic text extraction. ### [RivalSearchMCP](https://mcpcatalogs.com/en/server/damionrashford-rivalsearchmcp) by damionrashford · score 49 · ★89 RivalSearchMCP is a comprehensive research MCP server offering deterministic search across 5 web engines, 9 social platforms, academic databases, news sources, and document analysis. **When to choose:** Choose this when you need comprehensive research capabilities across multiple platforms without API keys or authentication, especially for competitive intelligence or multi-source fact-checking. **When NOT to choose:** Don't choose this if you need write access to databases or require highly specialized search that's not covered by its 10 available tools. ### [anysearch-mcp-server](https://mcpcatalogs.com/en/server/anysearch-ai-anysearch-mcp-server) by anysearch-ai · score 49 · ★221 Unified real-time search MCP server supporting web, domain-specific, batch search and content extraction. **When to choose:** Choose AnySearch when you need a comprehensive search solution with vertical domain specialization and batch processing capabilities. **When NOT to choose:** Avoid if you require search results with specific geographic filters not supported by the 23 available domains, or if you need offline search capabilities. ### [naver-search-mcp](https://mcpcatalogs.com/en/server/isnow890-naver-search-mcp) by isnow890 · score 48 · ★68 MCP server for Naver Search API integration with comprehensive search and DataLab analytics tools. **When to choose:** Choose this MCP server when you need comprehensive search capabilities across Naver's ecosystem and Korean market trend analysis. **When NOT to choose:** Don't choose this if you're only looking for general web search capabilities or if you don't need Korean-specific data and insights. ### [Kagi-Session2API-MCP](https://mcpcatalogs.com/en/server/ksroido-kagi-session2api-mcp) by KSroido · score 48 · ★134 Free Kagi Search MCP server using session tokens instead of paid API, works with multiple MCP clients. **When to choose:** Choose this when you want free access to Kagi's search and summarization features through MCP without paying for API calls. **When NOT to choose:** Don't choose if you risk violating Kagi's Terms of Service is a concern or if you need official API features like account balance tracking. ### [ServiceGraph](https://mcpcatalogs.com/en/server/nostrband-servicegraph) by nostrband · score 48 · ★160 An MCP server providing access to 100k+ US professional services firms with industry-specific search tools. **When to choose:** Choose this when you need to find specialized US professional services across multiple industries with filtering capabilities. **When NOT to choose:** Not suitable for international business needs as it only covers US firms, or when you need ongoing contact information beyond the monthly free tier. ### [MySearch-Proxy](https://mcpcatalogs.com/en/server/skernelx-mysearch-proxy) by skernelx · score 47 · ★120 Unified search MCP server that proxies requests to Tavily, Firecrawl, Exa and Social with a central management layer. **When to choose:** Choose MySearch-Proxy when you need a unified search interface for multiple AI assistants or want centralized management of search provider credentials and quotas. **When NOT to choose:** Avoid this if you need direct control over individual provider APIs without abstraction, or if you're not comfortable with the proxy-based architecture requiring additional infrastructure. ### [mcp-local-rag](https://mcpcatalogs.com/en/server/nkapila6-mcp-local-rag) by nkapila6 · score 47 · ★125 Local RAG web search MCP server with 9+ search backends and no API keys required. **When to choose:** Choose this when you need privacy-focused local web search without API dependencies and require multiple search backends for comprehensive research. **When NOT to choose:** Not recommended for production environments requiring high reliability or if you need web access that requires authentication. ### [server-google-news](https://mcpcatalogs.com/en/server/chanmeng666-server-google-news) by ChanMeng666 · score 47 · ★122 MCP server providing Google News search via SerpAPI with automatic categorization and multi-language support. **When to choose:** Choose this MCP server when you need reliable, categorized news search with multi-language support and are willing to work with SerpAPI's pricing model. **When NOT to choose:** Don't choose this if you need free unlimited news searches, as SerpAPI has usage limits and requires payment for higher volumes. ### [brave-search-mcp](https://mcpcatalogs.com/en/server/mikechao-brave-search-mcp) by mikechao · score 47 · ★117 MCP server integrating Brave Search API with web, image, video, news, local POI, and LLM context search capabilities. **When to choose:** Choose this when you need comprehensive search capabilities (web, images, news, local) integrated into your MCP-based AI workflows. **When NOT to choose:** Don't choose this if you need search results from multiple providers or have concerns about vendor lock-in with Brave's API. ### [gptr-mcp](https://mcpcatalogs.com/en/server/assafelovic-gptr-mcp) by assafelovic · score 47 · ★346 MCP server for deep research, providing comprehensive web search and report generation capabilities. **When to choose:** Choose GPT Researcher MCP when you need in-depth, high-quality research results rather than simple web search, especially when working with Claude Desktop or Docker-based deployments. **When NOT to choose:** Don't choose this server if you need rapid response times for simple queries, as the deep research process takes ~30 seconds, or if you require support for non-Python environments. ### [google-ai-mode-mcp](https://mcpcatalogs.com/en/server/pleaseprompto-google-ai-mode-mcp) by PleasePrompto · score 46 · ★117 MCP server enabling AI-powered Google AI Mode search with citations for any LLM client. **When to choose:** Choose this when you need free, efficient web research with AI-synthesized answers and citations, especially for coding, technical comparisons, and factual topics where source verification is important. **When NOT to choose:** Avoid this if you need real-time data, highly specialized content not covered by Google AI Mode, or if you're in a region where Google services are restricted without proxy configuration. ### [google-researcher-mcp](https://mcpcatalogs.com/en/server/zoharbabin-google-researcher-mcp) by zoharbabin · score 46 · ★35 MCP server providing Google search, web scraping, and document parsing with advanced filtering capabilities. **When to choose:** Choose this when you need comprehensive web research capabilities including Google search, news, academic papers, patents, and document parsing in a single MCP server. **When NOT to choose:** Avoid if you need to bypass Google's search restrictions or require functionality beyond what Google APIs provide, as this is tied to Google's services. ### [MetaSearchMCP](https://mcpcatalogs.com/en/server/gefsikatsinelou-metasearchmcp) by gefsikatsinelou · score 46 · ★44 A metasearch MCP server that aggregates results from multiple providers with normalized JSON output. **When to choose:** Choose this for comprehensive metasearch needs when you want to avoid vendor lock-in and need broad coverage across multiple search engines with a consistent JSON interface. **When NOT to choose:** Avoid this if you need real-time search results, as some providers use scraping that may be rate-limited, or if you require authentication for all search queries. ### [perplexity-web-api-mcp](https://mcpcatalogs.com/en/server/mishamyrt-perplexity-web-api-mcp) by mishamyrt · score 46 · ★72 An MCP server that provides access to Perplexity AI search and reasoning without requiring API keys by using session tokens. **When to choose:** Choose this when you want to access Perplexity's search capabilities without paying for their API, especially if you already have a Perplexity account. **When NOT to choose:** Avoid if you need high-volume commercial use or require stable API access, as it depends on web scraping that could break with UI changes. ### [qdrant-mcp-server](https://mcpcatalogs.com/en/server/mhalder-qdrant-mcp-server) by mhalder · score 46 · ★32 MCP server for semantic search using Qdrant vector database with multiple embedding providers. **When to choose:** Choose this when you need semantic search capabilities with local privacy-first vector storage or when working with existing Qdrant deployments. **When NOT to choose:** Avoid if you need support for other vector databases like Pinecone, Chroma, or Weaviate, as this is tightly coupled to Qdrant. ### [gnosis-mcp](https://mcpcatalogs.com/en/server/nicholasglazer-gnosis-mcp) by nicholasglazer · score 45 · ★22 Zero-config MCP server for searchable documentation with SQLite/PostgreSQL backends. **When to choose:** Choose Gnosis MCP when you need efficient local documentation search with strong performance metrics, zero cloud dependencies, and want to minimize token usage when AI agents query your documentation. **When NOT to choose:** Don't choose this if you need real-time collaborative features, write access to your documents, or require a hosted solution without local infrastructure management. ### [google-search](https://mcpcatalogs.com/en/server/web-agent-master-google-search) by web-agent-master · score 45 · ★595 Playwright-based Google search tool with MCP server integration that bypasses anti-bot detection. **When to choose:** Choose this when you need free, local Google search capabilities for AI assistants without paying for SERP APIs and want to maintain control over your data. **When NOT to choose:** Avoid if you need high-volume commercial search functionality, as this tool may get blocked with heavy usage; also not ideal if you can't run Node.js locally. ### [linkup-mcp-server](https://mcpcatalogs.com/en/server/linkupplatform-linkup-mcp-server) by LinkupPlatform · score 44 · ★28 Linkup MCP server provides web search and page fetching capabilities with real-time data and flexible depth options. **When to choose:** Choose Linkup MCP server when you need reliable, real-time web search capabilities and page fetching with advanced filtering options. **When NOT to choose:** Don't choose this if you're looking for an open-source solution, as it requires a proprietary API key from Linkup. ### [ReActMCP](https://mcpcatalogs.com/en/server/mshojaei77-reactmcp) by mshojaei77 · score 42 · ★143 ReActMCP is an MCP server that integrates web search capabilities with AI assistants using the Exa API. **When to choose:** Choose ReActMCP when you need real-time web search capabilities for an AI assistant, especially when you're already using the Exa API or want markdown-formatted search results with citations. **When NOT to choose:** Don't choose ReActMCP if you need a search solution without API dependency, require search functionality beyond what Exa API provides, or prefer an open-source search implementation. ### [context-lens](https://mcpcatalogs.com/en/server/cornelcroi-context-lens) by cornelcroi · score 42 · ★22 Semantic search knowledge base for MCP-enabled AI assistants that indexes local files or GitHub repos using LanceDB vector storage. **When to choose:** Choose Context Lens when you need semantic search capabilities for local files or GitHub repositories without setting up external services or cloud infrastructure. **When NOT to choose:** Avoid Context Lens if you need to search very large repositories (limited to 10MB per file), require collaborative features, or need to index non-textual file types. ### [academic-search-mcp-server](https://mcpcatalogs.com/en/server/afrise-academic-search-mcp-server) by afrise · score 41 · ★113 An MCP server that enables Claude to search and retrieve academic papers from Semantic Scholar and Crossref. **When to choose:** Choose this server if you're using Claude Desktop for academic research and need to search and retrieve information from scholarly databases. **When NOT to choose:** Don't choose if you need full-text access to papers (this only provides metadata and abstracts), or if you're not using Claude Desktop as your primary AI client. ### [bing-search-mcp](https://mcpcatalogs.com/en/server/leehanchung-bing-search-mcp) by leehanchung · score 40 · ★79 MCP server providing web, news and image search capabilities via Bing Search API. **When to choose:** Choose this server when you need Bing-specific search functionality and your AI assistant requires access to web, news, or image search capabilities. **When NOT to choose:** Don't choose this if you need search results from other providers like Google or DuckDuckGo, or if you require search capabilities beyond what Bing offers. ### [mcp-search](https://mcpcatalogs.com/en/server/caikiji-mcp-search) by caikiji · score 37 · ★1 MCP server for web search using SearXNG with privacy-respecting metasearch capabilities. **When to choose:** Choose this MCP server when you need privacy-respecting web search with control over your data and access to multiple search engines in one interface. **When NOT to choose:** Don't choose this if you need real-time search results from commercial engines like Google or Bing, or if you require features beyond what SearXNG provides. ### [mcp-servers-kagi](https://mcpcatalogs.com/en/server/ac3xx-mcp-servers-kagi) by ac3xx · score 36 · ★44 A TypeScript MCP server that integrates with Kagi's Search API to provide web search capabilities. **When to choose:** Choose this server if you need web search capabilities integrated into your AI workflow and are already using or planning to use Kagi's API. **When NOT to choose:** Don't choose this if you need alternative search engines, as it's specifically tied to Kagi's API which is currently in private beta. ### [Google-Search-MCP](https://mcpcatalogs.com/en/server/kongyo2-google-search-mcp) by kongyo2 · score 33 · ★3 Experimental MCP server for enhanced Google search with advanced anti-detection features. **When to choose:** Choose this server when you need Google search capabilities without access to official APIs and are willing to handle potential rate limiting and reliability issues. **When NOT to choose:** Don't choose this server for production use or if you need reliable, consistent search results, as it's experimental and may be blocked by Google's anti-scraping measures. ### [Glama-MCP-Server-Search](https://mcpcatalogs.com/en/server/kongyo2-glama-mcp-server-search) by kongyo2 · score 32 · ★2 An MCP server for searching and exploring MCP servers from the Glama directory. **When to choose:** Choose this MCP server when you need to discover and explore available MCP servers, particularly when you're building applications that integrate with multiple MCP services and need to understand the ecosystem landscape. **When NOT to choose:** Don't choose this if you only need to interact with a specific known MCP server or if you require functionality beyond directory exploration like direct tool access. ### [tmcp](https://mcpcatalogs.com/en/server/rbxwilliams-tmcp) by rbxwilliams · score 30 · ★0 JavaScript MCP server providing flight search tools with validation but no backend implementation. **When to choose:** Choose this when prototyping MCP servers for travel applications or testing MCP client implementations with a simple flight search interface. **When NOT to choose:** Don't choose this for production flight search as the backend is not implemented and the project appears to be in early development stage. ### [mcp-onesearch-api-poc](https://mcpcatalogs.com/en/server/bdobyns-mcp-onesearch-api-poc) by bdobyns · score 21 · ★0 Proof of concept MCP server implementing onesearch-api with various search tools and filters. **When to choose:** Choose this MCP server for prototyping search functionality within AI applications that need access to onesearch-api's specialized search capabilities. **When NOT to choose:** Avoid for production use as it's a proof of concept with no authentication and limited documentation. Consider established search MCP servers for critical applications. --- ## security (91) ### [ida-pro-mcp](https://mcpcatalogs.com/en/server/mrexodia-ida-pro-mcp) by mrexodia · score 60 · ★8,558 IDA Pro MCP server enabling AI-powered reverse engineering through MCP protocol integration. **When to choose:** Choose this server if you're performing reverse engineering tasks with IDA Pro and want to leverage AI assistance for code analysis and annotation. **When NOT to choose:** Avoid this if you don't have a licensed IDA Pro installation, need cross-platform support beyond what IDA offers, or require analysis capabilities for non-binary formats. ### [hexstrike-ai](https://mcpcatalogs.com/en/server/0x4m4-hexstrike-ai) by 0x4m4 · score 58 · ★8,764 HexStrike AI is an advanced MCP server that lets AI agents autonomously run 150+ cybersecurity tools for pentesting and security research. **When to choose:** Choose HexStrike AI if you need to integrate AI agents with offensive security capabilities, automate penetration testing workflows, or enhance bug bounty operations with intelligent tool selection. **When NOT to choose:** Avoid this if you're looking for defensive security tools, need integration with non-Python environments, or require support for compliance frameworks rather than offensive security testing. ### [ENScan_GO](https://mcpcatalogs.com/en/server/wgpsec-enscan-go) by wgpsec · score 56 · ★4,392 ENScan_GO is a Chinese enterprise information gathering tool that supports MCP server functionality. **When to choose:** Choose ENScan_GO when you need comprehensive Chinese enterprise intelligence and integration with AI assistants through MCP, especially for security research due diligence. **When NOT to choose:** Avoid using ENScan_GO if you need non-Chinese enterprise data, as it's specialized for Chinese business information sources only. ### [ThinkWatch](https://mcpcatalogs.com/en/server/thinkwatchproject-thinkwatch) by ThinkWatchProject · score 54 · ★868 Enterprise AI bastion host with unified proxying, RBAC, audit logs, rate limiting, and cost tracking across multiple LLM providers and MCP servers. **When to choose:** Choose ThinkWatch for enterprise environments requiring unified governance across multiple AI providers and MCP tools with comprehensive audit logging and cost tracking. **When NOT to choose:** Not suitable for simple personal use cases or teams that don't require enterprise-grade governance and audit trails. ### [ida-mcp-rs](https://mcpcatalogs.com/en/server/blacktop-ida-mcp-rs) by blacktop · score 54 · ★445 MCP server for AI-powered reverse engineering using IDA Pro, offering program analysis, decompilation, and scripting capabilities. **When to choose:** Choose this MCP server when you need AI assistance for reverse engineering tasks with IDA Pro and want to leverage conversational interfaces for binary analysis. **When NOT to choose:** Don't choose this if you don't have access to a licensed IDA Pro installation, as the server requires it to function. ### [jadx-ai-mcp](https://mcpcatalogs.com/en/server/zinja-coder-jadx-ai-mcp) by zinja-coder · score 54 · ★2,023 MCP server for JADX decompiler that enables AI-powered APK analysis with tools for reverse engineering and vulnerability detection. **When to choose:** Choose JADX-AI-MCP when you need to analyze Android APKs using AI assistance without leaving your decompiler environment, especially for reverse engineering and vulnerability detection tasks. **When NOT to choose:** Avoid if you need to analyze non-Android applications, as this tool is specifically designed for APK analysis only. ### [mcp-security-hub](https://mcpcatalogs.com/en/server/fuzzinglabs-mcp-security-hub) by FuzzingLabs · score 54 · ★548 A comprehensive collection of 38 production-ready MCP servers for offensive security tools including Nmap, Nuclei, SQLMap, and more. **When to choose:** Choose this collection when you need access to a comprehensive suite of offensive security tools through AI assistants, particularly for penetration testing, vulnerability assessment, and security research. **When NOT to choose:** Don't choose this if you're looking for defensive security tools, or if your use case requires tools not covered by the current 38 MCP servers in the collection. ### [LitterBox](https://mcpcatalogs.com/en/server/blacksnufkin-litterbox) by BlackSnufkin · score 53 · ★1,422 Red team payload sandbox with EDR integration and MCP tools for automated analysis. **When to choose:** Choose LitterBox when you need a comprehensive malware analysis platform with MCP integration for automated security testing, particularly for teams that already use multiple scanners and want a unified score for payload evaluation. **When NOT to choose:** Don't choose LitterBox if you need a cloud-based service, require extensive customization of analysis workflows beyond scanner integration, or need to analyze non-malware samples. ### [cve-mcp-server](https://mcpcatalogs.com/en/server/mukul975-cve-mcp-server) by mukul975 · score 53 · ★566 Production-grade MCP server providing Claude with 27 security intelligence tools across 21 APIs for CVE research and threat analysis. **When to choose:** Choose this server if you need comprehensive security intelligence capabilities including CVE analysis, threat intelligence, and network security tools in a single MCP server. **When NOT to choose:** Avoid if you need real-time threat detection capabilities or require integration with SIEM systems, as this is focused on intelligence gathering rather than active monitoring. ### [wassette](https://mcpcatalogs.com/en/server/microsoft-wassette) by microsoft · score 52 · ★888 Wassette is a security-oriented runtime that extends AI agents with WebAssembly components via MCP. **When to choose:** Choose Wassette when you need to execute potentially untrusted tools in a secure sandboxed environment or want to leverage the WebAssembly ecosystem for extending AI agents. **When NOT to choose:** Avoid Wassette if you need mature production-ready functionality or if you're not interested in leveraging the WebAssembly component ecosystem for your tools. ### [MCP-Kali-Server](https://mcpcatalogs.com/en/server/wh0am123-mcp-kali-server) by Wh0am123 · score 52 · ★688 MCP Kali Server connects AI agents to Linux systems for penetration testing and CTF challenges. **When to choose:** Choose this for AI-assisted penetration testing, CTF automation, or when you need to connect AI models directly to Kali Linux security tools. **When NOT to choose:** Don't use this if you need a broader security solution beyond penetration testing, or if you can't implement proper security measures for remote access. ### [pentest-ai](https://mcpcatalogs.com/en/server/0xsteph-pentest-ai) by 0xSteph · score 51 · ★251 Offensive-security MCP server with 205 wrapped tools, 17 specialist agents, and 60 SPA-aware probes for autonomous pentests. **When to choose:** Choose pentest-ai for automated offensive security testing through MCP clients when you need comprehensive OWASP Top 10 coverage with proof-of-concept generation. **When NOT to choose:** Avoid pentest-ai for production environments without proper authorization; its tools are designed for offensive security testing and require explicit permission. ### [h1-brain](https://mcpcatalogs.com/en/server/patrikfehrenbach-h1-brain) by PatrikFehrenbach · score 51 · ★300 MCP server connecting AI assistants to HackerOne for bug bounty hunting with personal data and public reports analysis. **When to choose:** Choose h1-brain if you're a security researcher doing bug bounty hunting on HackerOne and want AI assistance that combines your personal hunting history with public vulnerability knowledge. **When NOT to choose:** Don't choose h1-brain if you need support for vulnerability platforms other than HackerOne, or if you're uncomfortable sharing your HackerOne credentials through environment variables. ### [Wazuh-MCP-Server](https://mcpcatalogs.com/en/server/gensecaihq-wazuh-mcp-server) by gensecaihq · score 51 · ★174 MCP server enabling conversational queries to Wazuh SIEM for threat detection, incident response, and compliance checks. **When to choose:** Choose this if you're already using Wazuh SIEM and want to leverage AI for conversational security operations without replacing your existing security infrastructure. **When NOT to choose:** Don't choose this if you're not using Wazuh SIEM, as it requires a Wazuh deployment to function and won't integrate with other security platforms. ### [mcp-shodan](https://mcpcatalogs.com/en/server/burtthecoder-mcp-shodan) by BurtTheCoder · score 50 · ★127 MCP server for Shodan API providing IP reconnaissance, DNS lookups, and vulnerability intelligence. **When to choose:** Choose this MCP server when you need AI access to Shodan's network intelligence for security research, reconnaissance, or vulnerability intelligence gathering within your workflow. **When NOT to choose:** Don't use this if you don't have a Shodan API key, as it requires paid credits for most operations beyond basic lookups. ### [falcon-mcp](https://mcpcatalogs.com/en/server/crowdstrike-falcon-mcp) by CrowdStrike · score 50 · ★161 CrowdStrike's MCP server enables AI agents to access security operations through the Falcon platform. **When to choose:** Choose this when you need AI agents to interact with CrowdStrike Falcon's security platform for automated threat hunting and security analysis. **When NOT to choose:** Avoid this if you don't use CrowdStrike Falcon, need access to security tools from multiple vendors, or require production-ready features beyond public preview. ### [pentest-mcp](https://mcpcatalogs.com/en/server/dmontgomery40-pentest-mcp) by DMontgomery40 · score 50 · ★137 Professional penetration testing MCP server with nmap, JtR, hashcat, and reconnaissance tools. **When to choose:** Choose Pentest MCP for professional penetration testing when you need a comprehensive security testing toolkit with proper authentication and transport options for enterprise environments. **When NOT to choose:** Avoid this if you need basic security scanning without authentication, if you're in a restricted environment without ability to install required tools, or if you need testing against environments without explicit written permission. ### [ironcurtain](https://mcpcatalogs.com/en/server/provos-ironcurtain) by provos · score 50 · ★419 A secure runtime for AI agents with policy enforcement via natural language constitutions. **When to choose:** Choose IronCurtain when building autonomous AI agents that need strong security boundaries and policy enforcement, especially when working with sensitive operations like file system access, git operations, or API interactions. **When NOT to choose:** Avoid IronCurtain if you need high-performance computing with minimal overhead, as the security enforcement adds latency; or if you require deterministic behavior without human approval workflows. ### [mcp-server](https://mcpcatalogs.com/en/server/bitwarden-mcp-server) by bitwarden · score 50 · ★165 MCP server providing AI assistants with secure access to Bitwarden password manager vault management and organization administration features. **When to choose:** Choose this if you're using Bitwarden as your password manager and want your AI assistant to securely access and manage your vault and organization without switching tools. **When NOT to choose:** Don't choose this if you need to host the MCP server publicly, as it's designed exclusively for local use and would expose sensitive credentials. ### [agent-security-scanner-mcp](https://mcpcatalogs.com/en/server/sinewaveai-agent-security-scanner-mcp) by sinewaveai · score 49 · ★101 MCP security scanner for AI coding agents with prompt injection protection, package hallucination detection, vulnerability scanning, and auto-fix capabilities. **When to choose:** Choose this when building secure AI coding agents or autonomous assistants that need protection against code injection, prompt attacks, and supply chain vulnerabilities. **When NOT to choose:** Avoid if you need support for languages not in the 12-language list, or if you're looking for a solution without Python dependency (the advanced version requires Python). ### [aguara](https://mcpcatalogs.com/en/server/garagon-aguara) by garagon · score 49 · ★77 Security scanner for AI agent skills and MCP servers that detects prompt injection, data exfiltration, and supply-chain attacks before deployment. **When to choose:** Choose Aguara when you need to secure your AI applications by scanning MCP servers and agent skills for vulnerabilities before deployment. **When NOT to choose:** Don't choose Aguara if you're looking for an MCP server itself rather than a security tool for scanning MCP servers. ### [code-pathfinder](https://mcpcatalogs.com/en/server/shivasurya-code-pathfinder) by shivasurya · score 49 · ★130 MCP server for Code Pathfinder, providing AI assistants with access to cross-file code analysis and security scanning capabilities. **When to choose:** Choose this when you need AI assistants to have deep code security analysis capabilities beyond basic syntax checking, particularly for tracing data flows across files in complex applications. **When NOT to choose:** Don't choose this if you need support for languages other than Python, Docker, or Go, or if you require commercial support with SLAs as this is an open-source project without formal backing. ### [mcp-gateway](https://mcpcatalogs.com/en/server/lasso-security-mcp-gateway) by lasso-security · score 49 · ★371 MCP Gateway is a plugin-based intermediary that orchestrates multiple MCP servers with security features and monitoring capabilities. **When to choose:** Choose MCP Gateway when you need to manage multiple MCP servers with centralized security controls, especially for enterprise environments requiring token masking and guardrails against data leakage. **When NOT to choose:** Avoid if you need direct MCP server communication without intermediaries, or if you're looking for a lightweight solution with minimal overhead. ### [agentseal](https://mcpcatalogs.com/en/server/getagentseal-agentseal) by getagentseal · score 49 · ★256 Security toolkit for AI agents with MCP server scanning, prompt injection testing, and real-time monitoring capabilities. **When to choose:** Choose AgentSeal when you need to comprehensively secure your AI agent ecosystem, especially when deploying MCP servers or using third-party agents with potential security risks. **When NOT to choose:** Don't choose AgentSeal if you need continuous runtime protection rather than periodic scanning, or if you're looking for a solution that directly prevents attacks rather than detects them post-facto. ### [mcp-virustotal](https://mcpcatalogs.com/en/server/burtthecoder-mcp-virustotal) by BurtTheCoder · score 49 · ★125 MCP server for VirusTotal API providing comprehensive security analysis tools with automatic relationship data fetching. **When to choose:** Choose this server when you need comprehensive security analysis of URLs, files, IPs, or domains within your AI workflow, especially when automatic relationship data fetching is valuable. **When NOT to choose:** Avoid this if you need to analyze data from other security vendors or if you need more than the 1000 API requests per day free tier limit of VirusTotal. ### [mcp-server-wazuh](https://mcpcatalogs.com/en/server/gbrigandi-mcp-server-wazuh) by gbrigandi · score 49 · ★206 Rust-based MCP server for Wazuh SIEM enabling natural language queries to security alerts, vulnerabilities, and system metrics. **When to choose:** Choose this if you're already using Wazuh SIEM and want AI assistants to provide real-time security insights through natural language queries. **When NOT to choose:** Don't choose this if you need write capabilities to your SIEM or if you're not already invested in the Wazuh ecosystem. ### [cloudsword](https://mcpcatalogs.com/en/server/wgpsec-cloudsword) by wgpsec · score 49 · ★604 CloudSword is a Chinese cloud security tool that provides MCP protocol support since v0.0.2. **When to choose:** Choose CloudSword when you need a comprehensive cloud security assessment tool with MCP integration across multiple Chinese cloud providers. **When NOT to choose:** Avoid if you primarily work with non-Chinese cloud providers or need more specialized compliance frameworks not covered by this tool. ### [node9-proxy](https://mcpcatalogs.com/en/server/node9-ai-node9-proxy) by node9-ai · score 48 · ★180 Node9 is an execution security layer for AI agents that provides governance, monitoring, and audit logging through MCP integration. **When to choose:** Choose Node9 when you need enterprise-grade security governance for AI agents, especially if you work with sensitive systems and require audit trails of all agent actions. **When NOT to choose:** Don't choose Node9 if you need a lightweight solution with minimal overhead, or if you're only working with non-critical systems where comprehensive governance isn't required. ### [sysplant](https://mcpcatalogs.com/en/server/x42en-sysplant) by x42en · score 48 · ★130 SysPlant is a syscall hooking factory that provides multiple syscall retrieval methods with MCP server integration for AI coding assistants. **When to choose:** Choose SysPlant when you need to generate Windows syscall code with stealth capabilities for security research or penetration testing, and want to leverage MCP integration for direct generation within AI coding assistants. **When NOT to choose:** Don't use SysPlant for non-Windows systems, if you need support for languages other than NIM/C/C++/Rust, or if you're looking for a stable production tool with official support (it's research-focused with educational purposes only). ### [data-verify-mcp](https://mcpcatalogs.com/en/server/cccpan-data-verify-mcp) by CCCpan · score 48 · ★164 A comprehensive Chinese data verification MCP server offering 10 tools across 5 categories including identity verification, enterprise checks, vehicle info, OCR recognition, and risk assessment. **When to choose:** Choose this server when you need comprehensive Chinese data verification capabilities including identity checks, business information queries, vehicle verification, OCR recognition, and risk assessment for applications targeting Chinese markets. **When NOT to choose:** Don't choose this if you need verification services for non-Chinese data, require open source code implementation details, or need unlimited free usage beyond the 100 daily requests without registration. ### [vulnerable-mcp-servers-lab](https://mcpcatalogs.com/en/server/appsecco-vulnerable-mcp-servers-lab) by appsecco · score 48 · ★256 A collection of intentionally vulnerable MCP servers for security training and research into common attack vectors. **When to choose:** Security researchers and penetration testers who need practical, hands-on experience with MCP server vulnerabilities and attack vectors. **When NOT to choose:** Production environments or systems handling sensitive data, as these servers contain deliberate vulnerabilities that could be exploited. ### [BloodHound-MCP-AI](https://mcpcatalogs.com/en/server/mordavid-bloodhound-mcp-ai) by MorDavid · score 47 · ★353 MCP server connecting BloodHound with AI for natural language Active Directory security analysis. **When to choose:** Choose this if you're a security professional already using BloodHound who wants to analyze AD attack paths through natural language queries without learning Cypher. **When NOT to choose:** Don't choose this if you need real-time Active Directory analysis (this only works with collected data) or if you don't already have BloodHound data available. ### [nyxstrike](https://mcpcatalogs.com/en/server/commonhuman-lab-nyxstrike) by CommonHuman-Lab · score 47 · ★96 AI-powered offensive security orchestration engine with MCP integration for automated penetration testing. **When to choose:** Security researchers and penetration testers looking to automate attack chains with AI-driven tool orchestration in controlled environments. **When NOT to choose:** Users without proper security expertise or isolated environments, as NyxStrike gives AI direct access to offensive tools requiring specialized handling. ### [mcp-armor](https://mcpcatalogs.com/en/server/aira-security-mcp-armor) by aira-security · score 47 · ★115 MCP Armor is a security scanner that discovers, analyzes, and secures MCP servers integrated with various Agentic IDEs and clients. **When to choose:** Choose MCP Armor when you need to audit and secure MCP server configurations against security vulnerabilities and detect unauthorized changes in AI agent tool integrations. **When NOT to choose:** Don't choose MCP Armor if you need runtime enforcement and blocking capabilities, as it only performs static analysis without active protection mechanisms. ### [pfsense-mcp-server](https://mcpcatalogs.com/en/server/gensecaihq-pfsense-mcp-server) by gensecaihq · score 47 · ★69 pfSense MCP Server enables security administrators to manage firewalls through natural language commands with 327 tools and 9 safety layers. **When to choose:** Teams managing pfSense firewalls who want to reduce configuration complexity through natural language interfaces with built-in safety guardrails. **When NOT to choose:** You're using non-pfSense firewalls, as this MCP server is specifically designed for pfSense integration only. ### [ramparts](https://mcpcatalogs.com/en/server/highflame-ai-ramparts) by highflame-ai · score 47 · ★86 Ramparts is a security scanner for MCP servers that detects vulnerabilities and attack vectors in tool configurations. **When to choose:** Choose Ramparts when you need to validate the security posture of MCP servers before deployment, especially when connecting to third-party services or handling sensitive data through AI agents. **When NOT to choose:** Don't use Ramparts as your only security measure - it's designed for pre-deployment scanning and should be combined with runtime MCP guardrails for comprehensive protection. ### [mcp-auth-proxy](https://mcpcatalogs.com/en/server/sigbit-mcp-auth-proxy) by sigbit · score 47 · ★117 An OAuth 2.1 authentication proxy for MCP servers that adds security without code changes. **When to choose:** Choose mcp-auth-proxy when you need to add OAuth/OIDC authentication to one or a few MCP servers without modifying their code, especially for self-hosted or independently managed deployments. **When NOT to choose:** Don't choose this if you need to centrally manage multiple MCP servers with aggregation, catalog integration, or advanced policy enforcement - those features are in MCP Gateway instead. ### [mcp-gateway](https://mcpcatalogs.com/en/server/hyprmcp-mcp-gateway) by hyprmcp · score 47 · ★92 An enterprise-grade MCP gateway with OAuth authentication, prompt analytics, and firewall capabilities. **When to choose:** Choose this when building enterprise MCP deployments requiring centralized OAuth authentication and usage analytics without modifying existing MCP servers. **When NOT to choose:** Avoid if you need simple self-hosted authentication without telemetry, or if you prefer implementing OAuth directly in your MCP server for tighter integration. ### [x64dbg_mcp](https://mcpcatalogs.com/en/server/bromoket-x64dbg-mcp) by bromoket · score 46 · ★40 MCP server for x64dbg debugger with 23 mega-tools and 151 endpoints for reverse engineering with AI assistants. **When to choose:** Choose x64dbg MCP when you need AI assistance with reverse engineering tasks, particularly when working with x64dbg and want to leverage LLMs for automated debugging, analysis, and understanding of binary behavior. **When NOT to choose:** Avoid if you need debugger support beyond x64dbg, or if you prefer GUI-based debugging tools without AI integration. The server is specifically designed for Windows debugging and won't work with other platforms. ### [pentestMCP](https://mcpcatalogs.com/en/server/ramkansal-pentestmcp) by RamKansal · score 46 · ★65 MCP server exposing 20+ penetration testing tools for AI-assisted security testing via Docker. **When to choose:** Choose pentestMCP when you need AI-powered integration with comprehensive security assessment tools and are already using MCP-compatible environments like Claude Desktop. **When NOT to choose:** Don't choose pentestMCP if you need a GUI-based pentesting interface, have strict containerization requirements that conflict with its Docker setup, or need support for non-MCP AI platforms. ### [MCP-Defender](https://mcpcatalogs.com/en/server/mcp-defender-mcp-defender) by MCP-Defender · score 46 · ★252 Desktop app that scans and blocks malicious MCP traffic in AI apps like Cursor and Claude. **When to choose:** Choose MCP Defender when using AI applications that interact with MCP servers and you need protection against potentially malicious tool calls or data exfiltration. **When NOT to choose:** Don't choose MCP Defender if you need cross-platform support (currently only Mac), or if you prefer open-source security tools over commercial-acquired solutions. ### [mcp-guardian](https://mcpcatalogs.com/en/server/eqtylab-mcp-guardian) by eqtylab · score 46 · ★199 MCP Guardian is a security and management proxy for MCP servers, providing real-time control and monitoring. **When to choose:** Choose MCP Guardian when you need fine-grained control over MCP server interactions, particularly for compliance and security-sensitive environments. **When NOT to choose:** Don't choose MCP Guardian if you need maximum performance with minimal overhead, as it adds a proxy layer that may introduce latency to your MCP operations. ### [osint-mcp-server](https://mcpcatalogs.com/en/server/badchars-osint-mcp-server) by badchars · score 45 · ★17 An OSINT intelligence MCP server with 37 tools across 12 sources for comprehensive reconnaissance and attack surface mapping. **When to choose:** Choose this server when you need comprehensive OSINT capabilities for AI agents, especially when performing reconnaissance across multiple data sources simultaneously and correlating results automatically. **When NOT to choose:** Avoid this server if you need highly specialized OSINT tools not included in the 37 available tools, or if you require offline capabilities without API access to the included data sources. ### [mcpsec](https://mcpcatalogs.com/en/server/manthanghasadiya-mcpsec) by manthanghasadiya · score 45 · ★22 A security scanner and protocol fuzzer for MCP servers that discovers runtime vulnerabilities through dynamic testing. **When to choose:** Choose mcpsec when you need to validate the security posture of MCP servers before deployment or during development, especially when handling sensitive data or integrating with external systems. **When NOT to choose:** Don't choose mcpsec if you need a general-purpose security scanner or if you're looking for vulnerability assessment tools that support protocols other than MCP. ### [minibridge](https://mcpcatalogs.com/en/server/acuvity-minibridge) by acuvity · score 45 · ★54 Minibridge is a secure production bridge for MCP servers with policy enforcement, authentication, and monitoring capabilities. **When to choose:** Choose Minibridge when deploying MCP servers in production requiring security policies, integrity checks, and monitoring capabilities. **When NOT to choose:** Avoid for simple local development or when you need direct MCP server access without the overhead of a gateway. ### [MCP-Dandan](https://mcpcatalogs.com/en/server/82ch-mcp-dandan) by 82ch · score 45 · ★63 MCP-Dandan is a security monitoring framework that intercepts and analyzes MCP communications in real-time to detect malicious behavior. **When to choose:** Teams implementing MCP-based AI systems that require security monitoring, particularly those concerned about malicious tool usage or data exfiltration. **When NOT to choose:** Users needing a simple MCP integration without security features, or those who cannot accept the dual Python/Node.js dependency requirement. ### [MCPScan](https://mcpcatalogs.com/en/server/sahiloj-mcpscan) by sahiloj · score 45 · ★22 MCPScan is an offensive security auditor for MCP servers, detecting tool poisoning, credential leaks, RCE vectors, and supply chain vulnerabilities across stdio, HTTP, and SSE transports. **When to choose:** Choose MCPScan when you need to proactively audit MCP servers for security vulnerabilities before deployment or when integrating MCP into production systems. It's ideal for security teams and developers building MCP applications who want to identify potential risks. **When NOT to choose:** Don't choose MCPScan if you need a general-purpose security scanner for non-MCP systems or if you require protection against client-side attacks (MCPScan focuses on server-side vulnerabilities). ### [VibeShift](https://mcpcatalogs.com/en/server/groundng-vibeshift) by GroundNG · score 45 · ★67 VibeShift is an MCP server that provides automated security scanning and testing for AI-generated code. **When to choose:** Choose VibeShift when you need automated security scanning integrated directly into your AI coding workflow, especially for teams using static analysis tools like Semgrep or requiring automated UI testing with Playwright. **When NOT to choose:** Avoid VibeShift if you're looking for a comprehensive enterprise-grade security solution, as it's focused primarily on AI-generated code rather than full codebase analysis, or if you require advanced DAST capabilities beyond its basic implementation. ### [contrastapi](https://mcpcatalogs.com/en/server/upinar-contrastapi) by UPinar · score 45 · ★24 Security intelligence MCP server with 53 tools for CVE/KEV, threat intelligence, web scanning and code security analysis. **When to choose:** Choose ContrastAPI when you need comprehensive security intelligence tools for threat analysis, vulnerability assessment, and threat hunting without API key management. **When NOT to choose:** Not suitable for organizations requiring custom authentication models or those needing write capabilities to external systems, as this is a read-only security intelligence service. ### [js](https://mcpcatalogs.com/en/server/mcp-auth-js) by mcp-auth · score 44 · ★50 A Node.js SDK providing plug-and-play authentication for MCP servers using OAuth 2.1 and other RFC standards. **When to choose:** Choose MCP Auth when building Node.js MCP servers that need OAuth authentication and want to avoid implementing complex auth protocols from scratch. **When NOT to choose:** Don't choose this if you're not using Node.js, or if you need to implement custom authentication flows not supported by the SDK. ### [janee](https://mcpcatalogs.com/en/server/rsdouglas-janee) by rsdouglas · score 44 · ★28 Janee is an MCP server that provides secure secrets management for AI agents, enabling API access without exposing raw credentials. **When to choose:** Choose Janee when you're working with multiple AI agents that need API access but you want to maintain security through credential isolation and request auditing. **When NOT to choose:** Don't choose Janee if you need write access to services (currently read-only proxy only), or if you prefer a cloud-based secrets management solution rather than a local-first approach. ### [MCPHammer](https://mcpcatalogs.com/en/server/praetorian-inc-mcphammer) by praetorian-inc · score 44 · ★29 MCPHammer is a security testing framework that simulates vulnerable MCP servers for evaluation purposes. **When to choose:** Choose MCPHammer when you specifically need to test the security posture of MCP servers against injection attacks and remote execution risks. **When NOT to choose:** Avoid using MCPHammer for production environments or against servers you don't own. Its tools are designed for security testing and could cause harm if misused. ### [pentester-mcp](https://mcpcatalogs.com/en/server/halilkirazkaya-pentester-mcp) by halilkirazkaya · score 44 · ★34 MCP server integrating 200+ penetration testing tools for AI assistants with Docker sandbox security. **When to choose:** Choose Pentester-MCP when you need AI assistants to autonomously perform security assessments and penetration testing in a controlled, sandboxed environment. **When NOT to choose:** Avoid this if you need active directory write capabilities, as most tools are read-only reconnaissance utilities, or if you require testing in non-containerized environments. ### [Sentinelgate](https://mcpcatalogs.com/en/server/sentinel-gate-sentinelgate) by Sentinel-Gate · score 44 · ★25 SentinelGate provides MCP proxy with policy enforcement and auditing for AI agent actions. **When to choose:** Choose SentinelGate when you need to control and audit AI agent access to sensitive systems, especially when agents operate in containers or sandboxes with unrestricted tool permissions. **When NOT to choose:** Don't choose SentinelGate if you're looking for an MCP server to expose tools/resources to agents, as it's actually a security proxy that controls agent actions, not a provider of MCP capabilities. ### [bluerock](https://mcpcatalogs.com/en/server/bluerock-io-bluerock) by bluerock-io · score 44 · ★29 BlueRock provides runtime visibility for Python MCP servers with zero code changes, monitoring tool calls, sessions, imports and subprocess execution as structured NDJSON. **When to choose:** Choose BlueRock when you need security monitoring for MCP servers without modifying your application code, especially when you want visibility into tool calls, session lifecycle, and module integrity with SHA-256 verification. **When NOT to choose:** Don't choose BlueRock if you need Windows support (Linux/macOS only), require features beyond MCP monitoring and import tracking, or if you're looking for a lightweight solution with minimal resource overhead. ### [mcp-wireshark](https://mcpcatalogs.com/en/server/khuynh22-mcp-wireshark) by khuynh22 · score 44 · ★36 An MCP server that integrates Wireshark/tshark with AI tools for network analysis. **When to choose:** Choose this MCP server when you need to integrate Wireshark/tshark capabilities into AI workflows for network analysis, debugging, or security monitoring without leaving your chat interface. **When NOT to choose:** Don't choose this if you need advanced Wireshark features not exposed by the MCP server, or if you're working in an environment where installing system dependencies like tshark is restricted. ### [zebbern-kali-mcp](https://mcpcatalogs.com/en/server/zebbern-zebbern-kali-mcp) by zebbern · score 44 · ★27 MCP server for Kali Linux penetration testing, providing AI agents with access to 130+ security tools via Docker. **When to choose:** Choose this MCP server when you need AI agents to perform security testing tasks with access to a comprehensive Kali Linux toolkit in an isolated environment. **When NOT to choose:** Avoid this if you need persistent tool results beyond the container lifecycle, or if you require more specialized security tools not included in the Kali Linux distribution. ### [cve-search_mcp](https://mcpcatalogs.com/en/server/roadwy-cve-search-mcp) by roadwy · score 44 · ★100 An MCP server providing secure access to CVE-Search API for vulnerability information lookup. **When to choose:** Choose this server when building security-focused AI assistants or developer tools that need vulnerability data integration. **When NOT to choose:** Don't choose this if you need write access to vulnerability databases or require real-time alerting beyond what the CVE-Search API provides. ### [ramibot](https://mcpcatalogs.com/en/server/ramibotai-ramibot) by RamiBotAI · score 43 · ★20 RamiBot is a local-first AI security platform with MCP integration for red/blue team operations. **When to choose:** Choose RamiBot when you need AI assistance in security operations with real tool execution in controlled environments, especially for red/blue team workflows that require human oversight for sensitive operations. **When NOT to choose:** Don't choose RamiBot if you need cloud-based SaaS solutions without Docker dependency, or if you require rapid deployment without containerization overhead. ### [masquerade](https://mcpcatalogs.com/en/server/postralai-masquerade) by postralai · score 43 · ★76 Masquerade MCP is a privacy firewall for LLMs that redacts sensitive data from PDFs before processing. **When to choose:** Choose Masquerade when working with sensitive PDF documents that need AI analysis but cannot have raw personal information exposed to LLMs. **When NOT to choose:** Don't choose Masquerade if you need to process document formats other than PDFs or if you're concerned about dependency on the Tinfoil service. ### [pic-standard](https://mcpcatalogs.com/en/server/madeinplutofabio-pic-standard) by madeinplutofabio · score 43 · ★23 MCP server providing Provenance & Intent Contracts (PIC) to verify agent actions before execution. **When to choose:** Choose PIC when building AI systems that must verify the intent and provenance of high-impact actions before execution, particularly in financial, privacy, or regulated environments. **When NOT to choose:** Don't choose PIC if you need simple tool access controls without provenance verification, or if you're working in an environment where local-only solutions aren't acceptable. ### [rigour](https://mcpcatalogs.com/en/server/rigour-labs-rigour) by rigour-labs · score 43 · ★20 Rigour is an AI coding agent security system that detects secrets, structural issues, and AI drift through MCP tools. **When to choose:** Choose Rigour when you need real-time security and governance for AI-generated code, especially in teams where AI agents autonomously write code and need guardrails. **When NOT to choose:** Avoid Rigour if you need traditional code analysis without AI integration or if you require compliance with strict audit trails for all code changes. ### [a2a-demos](https://mcpcatalogs.com/en/server/capiscio-a2a-demos) by capiscio · score 43 · ★18 Demo agents showcasing CapiscIO Agent Guard and MCP Guard with trust badges, identity verification, and tool-level authorization. **When to choose:** When building secure AI agents or MCP servers that require identity verification, trust badges, and policy-based access control. **When NOT to choose:** When you don't need security features or prefer not to rely on CapiscIO's registry for trust infrastructure. ### [python](https://mcpcatalogs.com/en/server/mcp-auth-python) by mcp-auth · score 43 · ★57 A Python SDK providing plug-and-play OAuth 2.1 authentication for MCP servers. **When to choose:** Choose this when developing a Python MCP server and need to quickly implement OAuth 2.1 authentication without building auth infrastructure from scratch. **When NOT to choose:** Not ideal if you need non-Oauth authentication methods or are building servers in languages other than Python. ### [aws-security-mcp](https://mcpcatalogs.com/en/server/groovybugify-aws-security-mcp) by groovyBugify · score 43 · ★83 A MCP server connecting AI assistants to AWS security services for infrastructure analysis. **When to choose:** Choose this MCP server if you need AI-powered security analysis of AWS infrastructure through natural language queries and are already using AWS security services like GuardDuty and SecurityHub. **When NOT to choose:** Don't choose this if you need security analysis for non-AWS cloud providers, write access to resources (this is read-only), or if you're on a free tier of Claude (requires Pro/Max for larger token limits). ### [toolbox](https://mcpcatalogs.com/en/server/go-appsec-toolbox) by go-appsec · score 43 · ★24 MCP-based application security testing tools that enable collaboration between humans and coding agents via CLI and proxy. **When to choose:** Choose this MCP server when you need collaborative application security testing that combines human UI interaction with AI-powered traffic analysis and mutation. **When NOT to choose:** Don't choose this if you need standalone scanning without human involvement, or if you're already deeply integrated with Burp Suite and prefer its workflow. ### [shodan-mcp](https://mcpcatalogs.com/en/server/vorota-ai-shodan-mcp) by Vorota-ai · score 43 · ★19 An MCP server providing 20 tools for Shodan reconnaissance, vulnerability intelligence, and DNS analysis. **When to choose:** Choose this server when you need internet reconnaissance and vulnerability intelligence integrated directly into your AI workflow, especially for security research. **When NOT to choose:** Avoid this if you need active scanning capabilities (this is passive-only), have concerns about vendor lock-in to Shodan, or require more comprehensive vulnerability databases. ### [opnsense-mcp-server](https://mcpcatalogs.com/en/server/pixelworlds-opnsense-mcp-server) by Pixelworlds · score 42 · ★42 Modular MCP server for OPNsense firewall management with 88 tools accessing 2000+ methods. **When to choose:** Choose this if you're managing an OPNsense firewall and want AI assistants to help with configuration, monitoring, and troubleshooting through a well-structured, type-safe interface. **When NOT to choose:** Don't choose this if you need direct CLI access to OPNsense (bypassing AI assistants), are using other firewall platforms, or need features not exposed through the OPNsense API. ### [mcp-security-audit](https://mcpcatalogs.com/en/server/qianniuspace-mcp-security-audit) by qianniuspace · score 42 · ★52 MCP server for real-time npm security vulnerability scanning with detailed reports. **When to choose:** Choose this MCP server for Node.js projects needing real-time security scanning without complex setup. **When NOT to choose:** Not ideal for projects requiring deep custom rules or non-JavaScript ecosystems as it's focused solely on npm packages. ### [mitre-attack-mcp](https://mcpcatalogs.com/en/server/stoyky-mitre-attack-mcp) by stoyky · score 42 · ★40 MCP server providing 50+ tools for querying MITRE ATT&CK knowledge base with automatic Navigator layer generation. **When to choose:** Choose this server for cybersecurity analysis, threat intelligence, and ATT&CK matrix visualization when working with MITRE ATT&CK data. **When NOT to choose:** Avoid if you need access to non-MITRE threat intelligence sources or require real-time threat data beyond the periodically updated ATT&CK knowledge base. ### [mcp-server-idapro](https://mcpcatalogs.com/en/server/fdrechsler-mcp-server-idapro) by fdrechsler · score 41 · ★98 An MCP server bridging AI assistants with IDA Pro for reverse engineering tasks. **When to choose:** Choose this MCP server when you need AI assistance for reverse engineering tasks and are already using IDA Pro as your primary disassembler. **When NOT to choose:** Don't choose this if you don't have a license for IDA Pro, need to support other disassemblers, or require a more integrated reverse engineering platform. ### [mcp-auth](https://mcpcatalogs.com/en/server/famma-ai-mcp-auth) by famma-ai · score 41 · ★26 SDK for building OAuth-protected Remote MCP servers on Cloudflare Workers with pluggable auth adapters. **When to choose:** Choose this when you need to deploy an MCP server on Cloudflare Workers and require OAuth authentication, especially when your identity provider doesn't offer native OAuth 2.1 support. **When NOT to choose:** Don't choose this if you're not using Cloudflare Workers for deployment, or if your identity provider already has native OAuth 2.1 support that you can integrate directly. ### [agentidentityprotocol](https://mcpcatalogs.com/en/server/openagentidentityprotocol-agentidentityprotocol) by openagentidentityprotocol · score 41 · ★27 AIP provides zero-trust security layer for MCP agents with policy enforcement, human approval, and audit logging. **When to choose:** Choose AIP when you need to secure MCP deployments with granular access control for AI agents, especially in enterprise environments requiring audit trails and compliance. **When NOT to choose:** Don't choose AIP if you're looking for MCP servers that provide tools/resources directly, or if you need a simpler authentication mechanism for personal use cases. ### [awsome_kali_MCPServers](https://mcpcatalogs.com/en/server/ccq1-awsome-kali-mcpservers) by ccq1 · score 41 · ★101 Collection of MCP servers for Kali Linux providing security tools like Nmap, objdump, and Wireshark through Docker containers. **When to choose:** Choose this if you need to integrate Kali Linux security tools into your MCP workflow for penetration testing, reverse engineering, or network analysis. **When NOT to choose:** Not ideal for production environments requiring persistent storage or advanced authentication as this is primarily a development-focused tool collection. ### [pentest-mcp-server](https://mcpcatalogs.com/en/server/layesec006-pentest-mcp-server) by LayeSec006 · score 40 · ★18 A production-ready MCP server that enables AI agents to perform autonomous penetration testing on Linux systems via SSH with persistent tmux sessions. **When to choose:** Choose this MCP server when you need persistent session management for complex penetration testing workflows with interactive tools like Metasploit, reverse shells, and long-running scans. **When NOT to choose:** Avoid if you need cross-platform support beyond Linux, have strict network policies preventing SSH connections, or require non-persistent isolated command execution. ### [jadx-mcp-server](https://mcpcatalogs.com/en/server/qtty-jadx-mcp-server) by Qtty · score 40 · ★23 A pure-Java MCP server enabling AI assistants to analyze and decompile Android APK files using JADX. **When to choose:** Choose this MCP server when you need to integrate APK reverse engineering capabilities into AI workflows without installing additional system dependencies. **When NOT to choose:** Avoid this if you need real-time analysis, as JADX processing can be resource-intensive for large APKs, or if you require advanced static analysis beyond decompilation. ### [ExternalAttacker-MCP](https://mcpcatalogs.com/en/server/mordavid-externalattacker-mcp) by MorDavid · score 39 · ★76 ExternalAttacker-MCP provides automated attack surface mapping tools integrated through Model Context Protocol for security reconnaissance. **When to choose:** Choose this server for security reconnaissance and external attack surface mapping when you need to use natural language commands to automate security assessments. **When NOT to choose:** Avoid this tool if you need to scan systems without explicit permission, require internal network scanning capabilities, or need integration with other security workflows. ### [keycloak-model-context-protocol](https://mcpcatalogs.com/en/server/christophenglisch-keycloak-model-context-protocol) by ChristophEnglisch · score 39 · ★42 MCP server for Keycloak administration that enables AI-powered user and realm management through Claude Desktop. **When to choose:** Choose this server if you're using Keycloak for identity management and want to integrate user administration capabilities directly into AI workflows. **When NOT to choose:** Don't choose this if you need more advanced Keycloak features like group management, realm configuration, or integration with other identity solutions. ### [frida-mcp](https://mcpcatalogs.com/en/server/gindhar2112-frida-mcp) by Gindhar2112 · score 38 · ★7 Frida-MCP enables AI-driven automated Android dynamic analysis through MCP protocol to control Frida for streamlined reverse engineering. **When to choose:** Choose Frida-MCP when you need AI assistance with Android reverse engineering tasks and already have a rooted device with Magisk and zygisk-gadget installed. **When NOT to choose:** Avoid this if you don't have a rooted Android device or are unwilling to modify your device's security setup. ### [ciphertrust-manager-mcp-server](https://mcpcatalogs.com/en/server/sanyambassi-ciphertrust-manager-mcp-server) by sanyambassi · score 38 · ★8 MCP Server for Thales CipherTrust Manager enabling AI assistants to interact with cryptographic key management operations through the ksctl CLI. **When to choose:** Choose this server if you need to integrate AI assistants with Thales CipherTrust Manager for cryptographic key management operations. **When NOT to choose:** Avoid this server if you need support for non-Thales cryptographic solutions or if you require official vendor support. ### [mcp-governance-sdk](https://mcpcatalogs.com/en/server/ithena-one-mcp-governance-sdk) by ithena-one · score 38 · ★37 A governance SDK adding identity, RBAC, auditing and logging capabilities to MCP servers. **When to choose:** Choose this when building enterprise MCP applications requiring robust authentication, authorization, audit trails, and need to integrate with existing governance infrastructure. **When NOT to choose:** Avoid if you need a simple MCP server without enterprise governance features, or if you prefer open-source solutions without optional vendor lock-in to the Ithena platform. ### [asterisk-mcp-server](https://mcpcatalogs.com/en/server/winfunc-asterisk-mcp-server) by winfunc · score 38 · ★33 Asterisk MCP server provides security scanning tools for code snippets and codebases via the Asterisk API. **When to choose:** Choose this server if you need security scanning integration for your AI coding assistant and are already using or willing to integrate with the Asterisk API. **When NOT to choose:** Avoid if you need offline security scanning capabilities or prefer open-source solutions without vendor dependencies. ### [mcp-identity](https://mcpcatalogs.com/en/server/mustafabagdatli-git-mcp-identity) by mustafabagdatli-git · score 38 · ★2 MCP server middleware for per-request cryptographic user attestation with signature verification. **When to choose:** When your MCP server performs high-stakes operations where you need to prove exactly which user authorized each specific request. **When NOT to choose:** For low-risk operations where the overhead of key management isn't justified, or if your users aren't technically capable of managing a keypair file. ### [Sonicwall-MCP-Server](https://mcpcatalogs.com/en/server/gensecaihq-sonicwall-mcp-server) by gensecaihq · score 37 · ★5 A production-ready MCP server for analyzing SonicWall firewall logs with natural language queries and threat detection. **When to choose:** Choose this MCP server if you need to analyze SonicWall firewall logs through natural language queries and are already using SonicOS 7.x or 8.x devices. **When NOT to choose:** Don't choose this if you're using a different firewall vendor, as this is specifically designed for SonicWall devices only. ### [chucknorris](https://mcpcatalogs.com/en/server/pollinations-chucknorris) by pollinations · score 37 · ★60 An MCP server delivering jailbreak prompts to LLMs with dynamic schema adaptation techniques. **When to choose:** Security researchers and AI developers who need to test LLM vulnerabilities and understand potential jailbreak techniques through MCP. **When NOT to choose:** You should not use this for production applications, as it's designed solely for security research purposes and may trigger security alerts in production environments. ### [firmis-scanner](https://mcpcatalogs.com/en/server/firmislabs-firmis-scanner) by Firmislabs · score 36 · ★0 Firmis is a security scanner that detects malicious behavior in Claude Skills, MCP Servers, and other AI agent platforms. **When to choose:** Choose Firmis when you need security scanning for multiple AI agent platforms (MCP, Claude Skills, Codex plugins, etc.) and want deterministic rule-based analysis without requiring AI model inference. **When NOT to choose:** Don't choose Firmis if you need runtime monitoring of already-deployed agents (it focuses on pre-deployment scanning) or if you require AI-powered exploit analysis (which is a paid feature). ### [docs](https://mcpcatalogs.com/en/server/mcp-auth-docs) by mcp-auth · score 36 · ★5 MCP Auth provides plug-and-play authentication solution for MCP servers, supporting any OAuth 2.1 or OpenID Connect provider. **When to choose:** When you need documentation and guidance for implementing authentication in your MCP server. **When NOT to choose:** When you're looking for an actual MCP server implementation with built-in authentication. ### [GreenboneMCP](https://mcpcatalogs.com/en/server/matteocolazilli-greenbonemcp) by matteocolazilli · score 35 · ★2 MCP server connecting AI to Greenbone/OpenVAS vulnerability management via GMP. **When to choose:** Choose this server if you're using Greenbone/OpenVAS and want AI assistance to automate vulnerability scanning workflows without writing custom integration code. **When NOT to choose:** Don't choose this if you need a commercial vulnerability scanner integration, or if you're not already using Greenbone/OpenVAS as it's specifically tied to this ecosystem. ### [mcp-oauth2.1-server](https://mcpcatalogs.com/en/server/quantgeekdev-mcp-oauth2-1-server) by QuantGeekDev · score 35 · ★8 Reference MCP server implementing OAuth 2.1 authorization for secure access control. **When to choose:** Choose this when implementing OAuth 2.1 authentication for your MCP server, especially if you're already using AWS Cognito or Keycloak. **When NOT to choose:** Not suitable if you need a simple authentication solution or are using different identity providers not supported by this implementation. ### [mcp-scan](https://mcpcatalogs.com/en/server/mltrev23-mcp-scan) by mltrev23 · score 33 · ★0 MCP-Scan is a security tool that statically and dynamically scans MCP connections for vulnerabilities like prompt injection and tool poisoning. **When to choose:** Choose MCP-Scan when you need to validate MCP connections for security vulnerabilities or enforce runtime guardrails on MCP tool usage. **When NOT to choose:** Avoid if you don't want to share tool descriptions with external services (even with opt-out), or need advanced features beyond security scanning. ### [mcpaccess-auth0-express](https://mcpcatalogs.com/en/server/hivetrail-ai-mcpaccess-auth0-express) by hivetrail-ai · score 30 · ★0 Auth0 authentication middleware for Express-based MCP servers with JWT validation and role-based access control. **When to choose:** Choose this when building Express-based MCP servers that need enterprise-grade authentication with Auth0 integration and role-based access control. **When NOT to choose:** Don't choose this if you're not using Auth0 or building non-Express MCP servers, as it requires Auth0 integration and Express framework. --- ## social-media (5) ### [xiaohongshu-mcp](https://mcpcatalogs.com/en/server/xpzouying-xiaohongshu-mcp) by xpzouying · score 59 · ★13,616 MCP server for automating Xiaohongshu operations including content posting, commenting, and user interaction. **When to choose:** Choose this MCP server if you need to automate interactions with Xiaohongshu platform and prefer a self-hosted solution over the browser plugin alternative. **When NOT to choose:** Avoid this if you need write access to other social platforms, as it's specifically designed for Xiaohongshu only. Also consider the browser plugin version if you prefer zero configuration. ### [reddit-mcp](https://mcpcatalogs.com/en/server/arindam200-reddit-mcp) by Arindam200 · score 48 · ★285 MCP server providing access to Reddit data through PRAW API, with both read-only and authenticated functionality. **When to choose:** Choose this when you need AI-powered access to Reddit data for analysis, community management, or content strategy with both read and write capabilities. **When NOT to choose:** Avoid if you need access to private subreddits or require enterprise-level API rate limits, as this server relies on Reddit's standard API with its limitations. ### [mcp-server-weibo](https://mcpcatalogs.com/en/server/qinyuanpei-mcp-server-weibo) by qinyuanpei · score 45 · ★45 A robust MCP server providing Weibo social media data access with automatic authentication and both HTTP and stdio interfaces. **When to choose:** Choose this MCP server when you need real-time access to Weibo data for AI applications that require Chinese social media integration, especially when you need user profiles, trending topics, and engagement metrics. **When NOT to choose:** Don't use this server if you require write access to Weibo (this is read-only only) or if you need data from other Chinese social media platforms like Xiaohongshu or Douyin. ### [mcp-reddit](https://mcpcatalogs.com/en/server/adhikasp-mcp-reddit) by adhikasp · score 44 · ★400 MCP server for fetching and analyzing Reddit content with support for different post types. **When to choose:** Choose this server when you need to analyze Reddit content as part of your AI workflow, especially for trending topics or community sentiment analysis. **When NOT to choose:** Don't choose this if you need access to private subreddits or require advanced Reddit features not covered by the server's toolset. ### [opentwitter-mcp](https://mcpcatalogs.com/en/server/kamflowersthemacrogod-opentwitter-mcp) by kamflowersthemacrogod · score 36 · ★0 MCP server for accessing Twitter/X data with user profiles, tweet search, and follower tracking. **When to choose:** Choose this when you need Twitter/X data access through MCP for social media monitoring, trend analysis, or influencer tracking without complex API setup. **When NOT to choose:** Avoid if you need real-time Twitter access, official API features, or have concerns about third-party data reliability and privacy. --- ## sports (1) ### [fantasy-football-mcp-public](https://mcpcatalogs.com/en/server/derekrbreese-fantasy-football-mcp-public) by derekrbreese · score 45 · ★34 Yahoo Fantasy Football MCP server offering intelligent lineup optimization, draft assistance, and league management. **When to choose:** Choose this server if you play Yahoo Fantasy Football and want AI-powered lineup optimization, draft assistance, and league management tools integrated directly into your Claude Desktop workflow. **When NOT to choose:** Don't choose this server if you use non-Yahoo fantasy football platforms or if you prefer a more generic fantasy football management tool that supports multiple platforms. --- ## translation (1) ### [simplelocalize-mcp-server](https://mcpcatalogs.com/en/server/simplelocalize-simplelocalize-mcp-server) by simplelocalize · score 36 · ★0 MCP server that provides AI-powered tools for managing translations through the SimpleLocalize API. **When to choose:** Choose this server if you're using SimpleLocalize for translation management and want to integrate it directly into your AI-assisted development workflow. **When NOT to choose:** Avoid if you're not using SimpleLocalize as it only integrates with their specific translation management platform. --- ## travel (1) ### [fli](https://mcpcatalogs.com/en/server/punitarani-fli) by punitarani · score 57 · ★2,489 Google Flights MCP server providing programmatic flight search with two main tools for specific dates and flexible date range searches. **When to choose:** Choose Fli when you need reliable flight search capabilities through Google Flights with extensive filtering options and are comfortable with using a reverse-engineered API. **When NOT to choose:** Avoid Fli if you need an officially supported Google integration, or if you require enterprise-level reliability guarantees that a reverse-engineered solution can't provide. --- ## weather (3) ### [weather-mcp-server](https://mcpcatalogs.com/en/server/ezh0v-weather-mcp-server) by ezh0v · score 49 · ★244 A lightweight Go MCP server that provides real-time weather data to AI assistants. **When to choose:** Choose this server if you need simple, real-time weather data integration for AI assistants without complex configuration. **When NOT to choose:** Don't choose this server if you need extensive historical weather data, forecasts beyond current conditions, or integration with other weather services. ### [mcp-server-weather-js](https://mcpcatalogs.com/en/server/hideya-mcp-server-weather-js) by hideya · score 33 · ★13 A simple Node.js MCP server providing US weather forecasts and alerts via two tools. **When to choose:** Choose this server when you need a lightweight solution for basic US weather information in your AI agent workflow. **When NOT to choose:** Don't choose this if you need weather data outside the US, historical weather information, or more detailed meteorological data. ### [openweathermap-mcp-server](https://mcpcatalogs.com/en/server/abhayymishraa-openweathermap-mcp-server) by abhayymishraa · score 24 · ★0 MCP server providing weather data from OpenWeatherMap API with Bun runtime. **When to choose:** Choose this server for simple weather data integration when you're already using Bun as your runtime and need basic forecast functionality. **When NOT to choose:** Avoid if you need robust error handling, extensive weather features, or if you're not using Bun as your runtime environment. --- ## web-scraping (41) ### [fetch](https://mcpcatalogs.com/en/server/modelcontextprotocol-servers-fetch) by modelcontextprotocol · score 76 · ★85,748 · official An MCP server that fetches web content and converts HTML to markdown, allowing LLMs to read web pages. **When to choose:** Choose this server when you need your AI system to access and process current web content, especially when you need to convert HTML to clean markdown for LLM consumption. **When NOT to choose:** Avoid this server if you need to access internal network resources (security risk) or if you require JavaScript execution as this server only handles static HTML content. ### [Scrapling](https://mcpcatalogs.com/en/server/d4vinci-scrapling) by D4Vinci · score 62 · ★50,346 Web scraping framework with adaptive parsing and anti-bot capabilities, now integrated as an MCP server. **When to choose:** Choose Scrapling for adaptive web scraping when you need to handle frequently changing websites or require robust anti-bot bypass capabilities within your AI agent workflows. **When NOT to choose:** Don't choose Scrapling if you need to scrape websites that explicitly prohibit scraping or if you require support for JavaScript-heavy applications with complex client-side rendering. ### [firecrawl-mcp-server](https://mcpcatalogs.com/en/server/firecrawl-firecrawl-mcp-server) by firecrawl · score 60 · ★6,317 Official Firecrawl MCP Server brings web scraping, search, and page interaction capabilities to MCP-compatible AI agents. **When to choose:** When you need reliable web scraping, search, and interaction capabilities for your AI agent and are already using or willing to use Firecrawl's API. **When NOT to choose:** When you need self-contained scraping without external dependencies or when you're concerned about vendor lock-in with Firecrawl's API. ### [brightdata-mcp](https://mcpcatalogs.com/en/server/brightdata-brightdata-mcp) by brightdata · score 57 · ★2,366 A powerful MCP server providing real-time web access without blocking for AI agents. **When to choose:** Choose this MCP server when your AI agents need reliable, real-time web access without getting blocked, especially for research, data collection, or specialized knowledge extraction. **When NOT to choose:** Don't choose this if you need open-source self-hosting without vendor lock-in, as it requires an API token from BrightData and has limited free tier. ### [linkedin-mcp-server](https://mcpcatalogs.com/en/server/stickerdaniel-linkedin-mcp-server) by stickerdaniel · score 56 · ★1,900 MCP server providing LinkedIn API access through browser automation for profiles, companies, jobs, and messaging. **When to choose:** Choose this server when you need AI assistants to interact with LinkedIn data through a natural browser interface rather than unofficial APIs, especially for complex profile analysis and messaging tasks. **When NOT to choose:** Avoid this server if you need high-volume data processing, as LinkedIn may flag automated activity; also not suitable if you require programmatic access without browser automation. ### [apify-mcp-server](https://mcpcatalogs.com/en/server/apify-apify-mcp-server) by apify · score 55 · ★1,223 Apify MCP server enables AI agents to extract web data using thousands of pre-built scrapers and automation tools. **When to choose:** Choose Apify MCP server when your AI agent needs to extract structured data from websites at scale and you prefer using ready-made scrapers over building custom solutions. **When NOT to choose:** Avoid if you need fine-grained control over scraping behavior, have strict data privacy requirements, or prefer to manage your own infrastructure rather than relying on a third-party service. ### [webclaw](https://mcpcatalogs.com/en/server/0xmassi-webclaw) by 0xMassi · score 55 · ★1,155 A Rust-based fast local-first web content extraction tool with MCP server for AI agents. **When to choose:** Choose WebClaw when you need reliable, clean web content extraction for AI agents, especially if you prefer local-first processing and want multiple output formats (markdown, JSON, LLM-ready text). **When NOT to choose:** Don't choose WebClaw if you need JavaScript-heavy site rendering (requires cloud API) or if you need a different license than AGPL-3.0 for commercial redistribution. ### [XActions](https://mcpcatalogs.com/en/server/nirholas-xactions) by nirholas · score 51 · ★265 XActions is a comprehensive X/Twitter automation toolkit with MCP server integration for AI agents. **When to choose:** Choose XActions when you need comprehensive Twitter automation without API restrictions and want to integrate social media actions into AI agents via MCP. **When NOT to choose:** Don't choose XActions if you need stable, official API access or if you're concerned about potential changes to X's web interface breaking automation. ### [kindly-web-search-mcp-server](https://mcpcatalogs.com/en/server/shelpuk-ai-technology-consulting-kindly-web-search-mcp-server) by Shelpuk-AI-Technology-Consulting · score 50 · ★325 Web search MCP server with robust content retrieval for AI coding tools and agents. **When to choose:** Choose Kindly when you need comprehensive content retrieval beyond basic search results, especially for coding assistants that need full conversations, GitHub issues, or documentation rather than just URLs and snippets. **When NOT to choose:** Don't choose Kindly if you only need simple web search results, want to avoid installing a browser, or prefer solutions with fewer dependencies and API key requirements. ### [facebook-ads-library-mcp](https://mcpcatalogs.com/en/server/proxy-intell-facebook-ads-library-mcp) by proxy-intell · score 49 · ★225 Facebook Ads Library MCP server for querying and analyzing advertising campaigns with batch processing and video analysis capabilities. **When to choose:** Choose this server if you need to analyze Facebook ad campaigns at scale, especially when comparing multiple brands or performing video ad analysis. **When NOT to choose:** Don't choose this if you need to analyze ads from platforms other than Facebook, or if you can't/won't use third-party APIs requiring paid credits. ### [oxylabs-mcp](https://mcpcatalogs.com/en/server/oxylabs-oxylabs-mcp) by oxylabs · score 48 · ★95 Oxylabs MCP server enables AI models to scrape and interact with the real-world web. **When to choose:** Choose Oxylabs MCP when you need reliable access to web data with anti-bot bypass capabilities and structured output, especially if you already use or are willing to subscribe to Oxylabs services. **When NOT to choose:** Avoid Oxylabs MCP if you need open-source solutions, have concerns about vendor lock-in, or prefer not to pay for premium web scraping services. ### [agentql-mcp](https://mcpcatalogs.com/en/server/tinyfish-io-agentql-mcp) by tinyfish-io · score 48 · ★170 AgentQL MCP server enables structured web data extraction via a single tool with clear configuration across multiple MCP clients. **When to choose:** Choose this server when you need to extract structured data from complex web pages using natural language prompts, especially when working with AI agents that need real-time web data. **When NOT to choose:** Avoid this if you need to extract data from pages behind authentication walls, as it doesn't support login sessions, or if you're concerned about vendor lock-in to AgentQL's API. ### [mcp-read-website-fast](https://mcpcatalogs.com/en/server/just-every-mcp-read-website-fast) by just-every · score 48 · ★146 A fast, token-efficient MCP server that converts websites to clean Markdown for AI agents. **When to choose:** Choose this when you need fast, token-efficient web scraping for AI agents, especially when working with Claude Code or IDEs and need to convert websites to clean markdown. **When NOT to choose:** Don't choose this if you need to interact with JavaScript-heavy websites that require execution to render content, or if you need advanced authentication mechanisms beyond custom user agents. ### [scrapegraph-mcp](https://mcpcatalogs.com/en/server/scrapegraphai-scrapegraph-mcp) by ScrapeGraphAI · score 48 · ★72 ScrapeGraph MCP server provides AI-powered web scraping tools including extract, search, crawl, schema generation, and monitoring capabilities. **When to choose:** Choose this server when you need reliable, AI-powered web scraping capabilities through MCP and are already using or willing to integrate with the ScrapeGraph AI API. **When NOT to choose:** Don't choose this if you need open-source scraping without API dependency, want more control over the scraping process, or are concerned about vendor lock-in to ScrapeGraph's services. ### [apitap](https://mcpcatalogs.com/en/server/n1byn1kt-apitap) by n1byn1kt · score 48 · ★83 ApiTap turns any website into an API through MCP server with pre-mapped endpoints and traffic capture. **When to choose:** Choose ApiTap when you need API access to websites that don't have public APIs, and when you want to reduce token costs compared to browser automation. **When NOT to choose:** Don't choose ApiTap if you need write access to APIs (it's read-only), or if you're working with highly protected APIs that require complex request signatures or bot detection. ### [RedNote-MCP](https://mcpcatalogs.com/en/server/ifuryst-rednote-mcp) by iFurySt · score 47 · ★1,053 RedNote MCP server provides programmatic access to XiaoHongShu content through search and URL fetching. **When to choose:** Choose this server when you need to programmatically access RedNote content for analysis or research, especially if you're already using MCP-compatible clients like Cursor. **When NOT to choose:** Avoid if you need real-time data access or comprehensive comment functionality, as this server relies on cookie authentication that may expire and comment access is not yet implemented. ### [one-search-mcp](https://mcpcatalogs.com/en/server/yokingma-one-search-mcp) by yokingma · score 47 · ★114 OneSearch MCP Server provides web search, scraping, and content extraction with multiple search providers and local browser automation. **When to choose:** Choose OneSearch when you need an MCP server that combines multiple search engines with local browser automation capabilities, especially when you want to avoid API costs for basic scraping and search operations. **When NOT to choose:** Don't choose OneSearch if you need heavy scraping at scale, as browser automation may not be as efficient as specialized scraping tools for large-scale data extraction. ### [tradingview-chart-mcp](https://mcpcatalogs.com/en/server/ertugrul59-tradingview-chart-mcp) by ertugrul59 · score 47 · ★93 MCP server that captures TradingView chart images via Selenium with browser pooling for concurrent performance. **When to choose:** Choose this when you need high-performance concurrent chart scraping from TradingView with proper authentication and performance monitoring. **When NOT to choose:** Don't choose this if you need to scrape frequently updated charts without TradingView authentication or require non-chart data from TradingView. ### [crw](https://mcpcatalogs.com/en/server/us-crw) by us · score 47 · ★88 Fast, lightweight Firecrawl alternative in Rust with MCP server for AI agents. Web scraper, crawler & search API. **When to choose:** Choose CRW when you need a fast, lightweight web scraping solution for AI agents that can be self-hosted with minimal resources or accessed via cloud API with Firecrawl compatibility. **When NOT to choose:** Avoid CRW if you need commercial-level support and SLA guarantees, require advanced JavaScript rendering beyond Chrome capabilities, or need to handle extremely large-scale scraping operations beyond what a single binary can efficiently process. ### [token-enhancer](https://mcpcatalogs.com/en/server/xelektron-token-enhancer) by xelektron · score 46 · ★64 MCP server that strips web pages to clean text, reducing token usage by 86-99%. **When to choose:** Choose this when you need to significantly reduce token consumption when processing web content in AI workflows, especially when working with data-heavy sites like financial reports or documentation. **When NOT to choose:** Don't choose this if you need JavaScript-rendered content processing, as it only handles static HTML, JSON, and plain text. ### [x-twitter-scraper](https://mcpcatalogs.com/en/server/xquik-dev-x-twitter-scraper) by Xquik-dev · score 46 · ★82 MCP server with 2 tools for Twitter/X data extraction and API interaction via 100+ endpoints. **When to choose:** Choose this for affordable access to Twitter data when you need bulk extraction, monitoring, and posting capabilities without using the official API. **When NOT to choose:** Don't use if you need official API compliance, have concerns about X's terms of service, or require stability guarantees for production workloads. ### [crawl4ai-mcp-server](https://mcpcatalogs.com/en/server/sadiuysal-crawl4ai-mcp-server) by sadiuysal · score 46 · ★85 A lightweight MCP server exposing Crawl4AI web scraping tools for AI agents with Docker support. **When to choose:** Choose this when you need a self-hosted, free alternative to commercial web scraping APIs like Firecrawl, especially when working with sensitive data or requiring offline capabilities. **When NOT to choose:** Avoid this if you need advanced OCR capabilities for image-heavy sites or require official enterprise support with service-level agreements. ### [mcp-llms-txt-explorer](https://mcpcatalogs.com/en/server/thedaviddias-mcp-llms-txt-explorer) by thedaviddias · score 45 · ★74 MCP server for discovering and analyzing websites with llms.txt files. **When to choose:** Choose this server if you need to discover, validate, or work with websites that implement the llms.txt standard for AI access. **When NOT to choose:** Don't choose this server if you need to work with websites that don't follow the llms.txt standard or if you need more advanced web scraping capabilities beyond simple file discovery. ### [crawlbase-mcp](https://mcpcatalogs.com/en/server/crawlbase-crawlbase-mcp) by crawlbase · score 45 · ★55 Crawlbase MCP Server connects AI agents to live web data through structured scraping, JavaScript rendering, and anti-bot protection. **When to choose:** Choose this server when you need reliable, real-time web data in your AI workflows and want battle-tested scraping infrastructure with anti-bot protection. **When NOT to choose:** Avoid if you need unlimited free usage, as Crawlbase requires tokens that eventually need a paid plan, or if you prefer self-hosted solutions without vendor lock-in. ### [html-to-markdown-mcp](https://mcpcatalogs.com/en/server/levz0r-html-to-markdown-mcp) by levz0r · score 45 · ★24 MCP server that converts HTML to Markdown using Turndown.js with SSRF protection. **When to choose:** Choose this MCP server when you need to convert HTML from web pages or raw HTML content to clean Markdown with proper formatting. **When NOT to choose:** Don't choose this if you need to process complex JavaScript-heavy sites that require browser rendering, as it only fetches static HTML. ### [decipher-research-agent](https://mcpcatalogs.com/en/server/mtwn105-decipher-research-agent) by mtwn105 · score 45 · ★151 AI-powered research assistant that uses MCP server to create interactive notebooks from web sources. **When to choose:** Choose DecipherIt when you need a comprehensive research platform that combines document analysis, web scraping, and AI-powered content synthesis into a single application. **When NOT to choose:** Don't choose DecipherIt if you're looking for an MCP server to integrate with other applications, as this is a standalone research assistant application that consumes MCP services. ### [mcp-wayback-machine](https://mcpcatalogs.com/en/server/mearman-mcp-wayback-machine) by Mearman · score 43 · ★24 MCP server for Internet Archive's Wayback Machine with CDX search, snapshot retrieval, comparison and optional authentication. **When to choose:** Choose this MCP server when you need to archive, retrieve, or compare web pages through AI agents, especially when working with historical web content or tracking changes to websites over time. **When NOT to choose:** Avoid if you need to upload files to the Internet Archive (this only works with the Wayback Machine) or if you require non-commercial use only due to the CC BY-NC-SA license. ### [builtwith-api](https://mcpcatalogs.com/en/server/zcaceres-builtwith-api) by zcaceres · score 43 · ★20 A robust MCP server that exposes 13 BuiltWith API tools for web technology analysis via LLM agents. **When to choose:** Choose this when you need AI agents to analyze web technologies, track tech trends, or understand competitor technology stacks without manual research. **When NOT to choose:** Don't choose this if you need real-time monitoring or modification of websites, as this is a read-only analysis tool. ### [mcp-webresearch](https://mcpcatalogs.com/en/server/mzxrai-mcp-webresearch) by mzxrai · score 41 · ★297 MCP server enabling Claude to perform web research with Google search, page extraction, and screenshot capture capabilities. **When to choose:** Choose this MCP server when you need real-time web research capabilities directly within Claude, particularly for current events or topics requiring up-to-date information from the web. **When NOT to choose:** Avoid if you need reliable, production-ready code (marked as pre-alpha) or require support for platforms beyond macOS. ### [webfetch-mcp](https://mcpcatalogs.com/en/server/manooll-webfetch-mcp) by manooll · score 41 · ★26 An MCP server providing real-time web access with tunable search and clean content extraction for local AI models. **When to choose:** Choose WebFetch.MCP when you need privacy-focused web access for local AI models without API keys, requiring fine-tuned search capabilities across technical and academic sources. **When NOT to choose:** Don't choose this if you need JavaScript-heavy content extraction (requires JSDOM) or require real-time content that needs immediate processing, as it has rate limiting (12 calls per 5 minutes). ### [Custom-MCP-Server](https://mcpcatalogs.com/en/server/sharan-kumar-r-custom-mcp-server) by Sharan-Kumar-R · score 40 · ★75 An MCP server providing scraping tools for LinkedIn, Facebook, Instagram, and Google search. **When to choose:** Choose this server if you need to scrape public social media profiles and perform Google searches through MCP, and already have RapidAPI and Serper.dev API keys. **When NOT to choose:** Avoid this server if you need to scrape non-public data, don't want to deal with third-party API limits, or require more advanced search capabilities than basic Google results. ### [scraperapi-mcp](https://mcpcatalogs.com/en/server/scraperapi-scraperapi-mcp) by scraperapi · score 39 · ★4 ScraperAPI MCP server enables LLMs to retrieve and process web scraping requests using ScraperAPI. **When to choose:** Choose this server when you need to scrape dynamic websites with JavaScript rendering, bypass geo-restrictions, or access premium residential proxies through your LLM client. **When NOT to choose:** Avoid this if you need a free scraping solution, as ScraperAPI is a paid service, or if you're concerned about vendor lock-in to a specific scraping infrastructure. ### [mcp-server-webscan](https://mcpcatalogs.com/en/server/bsmi021-mcp-server-webscan) by bsmi021 · score 37 · ★12 MCP server for web scanning with page fetching, link extraction, crawling, and sitemap generation. **When to choose:** Choose this MCP server when you need to analyze web content, perform site audits, or generate sitemaps through AI agents. **When NOT to choose:** Don't choose this if you need real-time monitoring, have strict performance requirements, or require authentication for accessing protected content. ### [mcp-server-requests](https://mcpcatalogs.com/en/server/coucya-mcp-server-requests) by coucya · score 37 · ★7 MCP server for HTTP requests with web fetching, content filtering, and markdown conversion capabilities. **When to choose:** When you need an MCP server for making HTTP requests with various content processing options and flexible workspace integration. **When NOT to choose:** If you need advanced web scraping with JavaScript rendering, complex rate limiting, or require more than basic authentication methods. ### [kakuyomu-mcp](https://mcpcatalogs.com/en/server/106-kakuyomu-mcp) by 106- · score 34 · ★2 An MCP server for accessing Japanese novel content from Kakuyomu, offering search, episode listing, and content retrieval tools. **When to choose:** Choose this server if you need to access Japanese web novels from Kakuyomu within an AI agent workflow for analysis, summarization, or content discovery. **When NOT to choose:** Don't use this if you need access to premium Kakuyomu content that requires login, as the server only scrapes public pages and may break if Kakuyomu changes their HTML structure. ### [brightdata-mcp-adk-hackathon](https://mcpcatalogs.com/en/server/arjunprabhulal-brightdata-mcp-adk-hackathon) by arjunprabhulal · score 34 · ★1 Professional web scraping platform using BrightData MCP server with 50+ tools and Google ADK integration. **When to choose:** Choose this if you need comprehensive web scraping capabilities with specialized tools for various data sources and already use or are willing to integrate with BrightData's proxy network. **When NOT to choose:** Don't choose this if you need an open-source solution without vendor lock-in, or if you only need basic scraping capabilities without the specialized tools provided by BrightData. ### [jinaai-mcp-server](https://mcpcatalogs.com/en/server/cyanheads-jinaai-mcp-server) by cyanheads · score 33 · ★0 A production-ready MCP server that enables AI systems to read and extract clean content from webpages using the Jina AI Reader API. **When to choose:** When you need a reliable, production-grade solution for extracting clean content from web pages and integrating it into AI workflows, especially if you're already using Jina AI's services. **When NOT to choose:** If you need to avoid vendor lock-in to Jina AI or if you require web scraping capabilities without external API dependencies. ### [EVE-University-Wiki-MCP-Server](https://mcpcatalogs.com/en/server/kongyo2-eve-university-wiki-mcp-server) by kongyo2 · score 33 · ★1 MCP server providing access to EVE University Wiki with Wayback Machine fallback. **When to choose:** Choose this MCP server when you need reliable access to EVE Online game knowledge, especially if you're concerned about wiki downtime due to the built-in Wayback Machine fallback. **When NOT to choose:** Don't choose this if you need access to gaming content outside of EVE Online, as the server is specifically dedicated to the EVE University Wiki. ### [hacker-mcp-server-POC](https://mcpcatalogs.com/en/server/sumitsk-simform-hacker-mcp-server-poc) by sumitsk-simform · score 33 · ★1 A lightweight MCP server exposing Hacker News data as tools for retrieving stories, details, and user profiles. **When to choose:** Choose this server when you need to integrate Hacker News data into AI workflows without complex setup, especially for prototyping or personal projects. **When NOT to choose:** Don't choose this for production systems as it lacks proper error handling, authentication, and has no rate limiting to protect against API throttling. ### [mcp-server-indexnow](https://mcpcatalogs.com/en/server/zizzfizzix-mcp-server-indexnow) by zizzfizzix · score 30 · ★0 An MCP server for submitting URLs to search engines via the IndexNow protocol. **When to choose:** Choose this server if you need to submit URLs to search engines for faster indexing and you're already using MCP-compatible clients like Claude Desktop or Zed. **When NOT to choose:** Don't choose this if you need bulk URL processing capabilities, as it appears designed for individual submissions rather than large-scale operations. ### [remote-mcp-server-authless-andor](https://mcpcatalogs.com/en/server/elizabethsiegle-remote-mcp-server-authless-andor) by elizabethsiegle · score 28 · ★1 Remote MCP server for web scraping using Cloudflare Browser Rendering and Workers AI, with no authentication required. **When to choose:** Choose this when you need a quickly deployable, auth-less MCP server for web scraping with AI capabilities, especially if you're already using Cloudflare services. **When NOT to choose:** Avoid this for production use requiring authentication or higher security needs, as it's designed without authentication and may not be suitable for sensitive data processing. ---