CoexistAI
by SPThole·★ 481·Score 51
Modular research assistant framework with MCP server integration for web search, YouTube, Reddit, mapping, and GitHub exploration.
Overview
CoexistAI is a comprehensive research assistant framework that provides modular tools for web exploration, social media content analysis, video summarization, location mapping, and codebase investigation. The framework supports multiple LLM providers including OpenAI, Google Gemini, and local models through Ollama. It can be deployed via Docker or local setup with both FastAPI endpoints and MCP server capabilities, making it highly versatile for different integration scenarios.
Try asking AI
After installing, here are 7 things you can ask your AI assistant:
When to choose this
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 this
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.
Tools this server exposes
12 tools extracted from the READMEweb-searchPOST /web-searchSearch the web, summarize results, and extract context using LLMs.
web-summarizePOST /web-summarizeSummarize any article or research paper by URL.
youtube-searchPOST /youtube-searchSearch YouTube videos and get summaries or answers.
reddit-searchPOST /reddit-searchSearch Reddit with custom phrases, sort options, and get top comments.
map-searchPOST /map-searchFind places, routes, and nearby points of interest.
git-tree-searchPOST /git-tree-searchGet the directory structure of any GitHub or local repository.
git-searchPOST /git-searchFetch, search, and analyze code in any GitHub or local repository.
text-to-podcastConvert text content into podcast episodes.
text-to-speechConvert text to high-quality audio using advanced TTS.
file-explorerExplore local folders and files with vision support for images.
llm-queryQuery various LLM models with customizable parameters.
embedding-generatorGenerate embeddings for text using various embedder models.
Comparable tools
Installation
Installation
**Method 1: Docker (Recommended)** Follow instructions in [README.docker.md](README.docker.md) for containerized setup.
**Method 2: Local Setup**
- Clone the repository:
git clone https://github.com/SPThole/CoexistAI.git coexistai && cd coexistai - Configure model settings in
config/model_config.json - Run setup script:
- macOS/Linux zsh: zsh quick_setup.sh - Linux bash: bash quick_setup.sh
**MCP Configuration for Claude Desktop** Add to Claude Desktop config.json:
{
"mcpServers": {
"coexistai": {
"command": "python",
"args": ["-m", "coexistai.app"]
}
}
}FAQ
- Which LLM providers are supported?
- OpenAI, Google Gemini, Ollama (local models), and other OpenAI-compatible providers like Groq and OpenRouter.
- Can I use CoexistAI without MCP?
- Yes, it can be used via FastAPI endpoints, Python library, or directly as a standalone application.
Compare CoexistAI with
Last updated · Auto-generated from public README + GitHub signals.