MCP Catalogs
Home

anysearch-mcp-server

by anysearch-ai·221·Score 49

Unified real-time search MCP server supporting web, domain-specific, batch search and content extraction.

searchai-llmproductivity
27
Forks
0
Open issues
this month
Last commit
2d ago
Indexed

Overview

AnySearch MCP Server provides a comprehensive search capability for AI agents, offering general web search, structured queries across 23 vertical domains, parallel batch search functionality, and full-page URL content extraction. The server supports multiple transport protocols including native Streamable HTTP, as well as SSE and stdio via proxies. It works both with and without API keys, providing anonymous access with lower rate limits when no key is configured.

Try asking AI

After installing, here are 5 things you can ask your AI assistant:

you:Real-time information retrieval for AI assistants across multiple domains
you:Batch processing multiple search queries simultaneously for efficiency
you:Content analysis and summarization from specific web pages or documents
you:Do I need an API key to use AnySearch?
you:Which transport protocols are supported?

When to choose this

Choose AnySearch when you need a comprehensive search solution with vertical domain specialization and batch processing capabilities.

When NOT to choose this

Avoid if you require search results with specific geographic filters not supported by the 23 available domains, or if you need offline search capabilities.

Tools this server exposes

4 tools extracted from the README
  • searchquery: string, domain?: string, sub_domain?: string, sub_domain_params?: object, content_types?: string[], zone?: string, max_results?: integer, freshness?: string

    Execute a search query — general or vertical domain

  • list_domainsdomain?: string, domains?: string[]

    Query the vertical domain directory to discover available sub_domains and their query formats

  • batch_searchqueries: object[]

    Execute 2-5 independent search queries in parallel

  • extracturl: string

    Fetch full page content from a URL and return as Markdown

Comparable tools

semantic-search-mcpperplexity-mcpweb-search-mcpserper-mcp

Installation

Streamable HTTP (Recommended)

For Claude Desktop (2025.6+):

{
  "mcpServers": {
    "anysearch": {
      "type": "streamable-http",
      "url": "https://api.anysearch.com/mcp",
      "headers": {
        "Authorization": "Bearer ${ANYSEARCH_API_KEY}"
      }
    }
  }
}

For stdio transport (using mcp-remote):

{
  "mcpServers": {
    "anysearch": {
      "command": "npx",
      "args": [
        "-y",
        "mcp-remote",
        "https://api.anysearch.com/mcp",
        "--header",
        "Authorization: Bearer ${ANYSEARCH_API_KEY}"
      ]
    }
  }
}

FAQ

Do I need an API key to use AnySearch?
No, an API key is optional. Without a key, you can still use all features via anonymous access with lower rate limits.
Which transport protocols are supported?
AnySearch natively supports Streamable HTTP (2025-03-26 spec), and supports SSE and stdio via proxy servers like mcp-remote or supergateway.

Compare anysearch-mcp-server with

GitHub →

Last updated · Auto-generated from public README + GitHub signals.