MCP Catalogs
Home

one-search-mcp

by yokingma·114·Score 47

OneSearch MCP Server provides web search, scraping, and content extraction with multiple search providers and local browser automation.

web-scrapingsearchbrowser-automation
17
Forks
2
Open issues
1 mo ago
Last commit
2d ago
Indexed

Overview

OneSearch MCP Server is a comprehensive web search and scraping tool that integrates with multiple search engines including SearXNG, Tavily, DuckDuckGo, Bing, and more. It features both API-based search and local browser automation through agent-browser, offering flexibility between free local operations and API-powered services. The server provides four main tools: one_search for web searching, one_scrape for content extraction, one_map for browser-based searches, and one_extract for content preprocessing. The implementation has recently transitioned from Firecrawl to agent-browser, improving privacy by handling browser operations locally without external API dependencies.

Try asking AI

After installing, here are 7 things you can ask your AI assistant:

you:Conduct web searches with multiple search providers through MCP
you:Scrape and extract content from websites for analysis
you:Perform local browser searches without API keys or costs
you:Preprocess web content for downstream AI model consumption
you:What browsers are supported by OneSearch MCP?
you:Do I need API keys for all search providers?
you:How do I migrate from v1.1.0 or earlier?

When to choose this

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 this

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.

Tools this server exposes

4 tools extracted from the README
  • one_search

    Web search with multiple search engines

  • one_scrape

    Scrape content from web pages

  • one_map

    Map web content using browser automation

  • one_extract

    Preprocess multi-URL page content for analysis

Comparable tools

mcp-firecrawlmcp-searxngweb-scraper-mcpbrave-search-mcp

Installation

Installation

Using Claude Desktop

Add to your Claude Desktop configuration file:

**macOS**: ~/Library/Application Support/Claude/claude_desktop_config.json **Windows**: %APPDATA%\Claude\claude_desktop_config.json

{
  "mcpServers": {
    "one-search-mcp": {
      "command": "npx",
      "args": ["-y", "one-search-mcp"],
      "env": {
        "SEARCH_PROVIDER": "local"
      }
    }
  }
}

Manual Installation

# Install globally (Optional)
npm install -g one-search-mcp

# Or run directly with npx
npx -y one-search-mcp

Using Docker

# Pull the image
docker pull ghcr.io/yokingma/one-search-mcp:latest

Configure with Docker in your Claude Desktop config:

{
  "mcpServers": {
    "one-search-mcp": {
      "command": "docker",
      "args": ["run", "-i", "--rm", "ghcr.io/yokingma/one-search-mcp:latest"],
      "env": {
        "SEARCH_PROVIDER": "local"
      }
    }
  }
}

FAQ

What browsers are supported by OneSearch MCP?
OneSearch MCP supports Chromium-based browsers including Google Chrome, Microsoft Edge, Chromium, and Google Chrome Canary. The server will automatically detect which browser is available.
Do I need API keys for all search providers?
No. Some providers like 'local' and 'duckduckgo' don't require API keys. Others like 'tavily', 'bing', 'google', 'zhipu', 'exa', and 'bocha' do require API keys.
How do I migrate from v1.1.0 or earlier?
Since v1.1.0 has removed Firecrawl integration, you need to install a Chromium browser and remove FIRECRAWL_API_URL and FIRECRAWL_API_KEY from your environment variables. Then update to the latest version.

Compare one-search-mcp with

GitHub →

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