MCP Catalogs
Home

google-search

by web-agent-master·595·Score 45

Playwright-based Google search tool with MCP server integration that bypasses anti-bot detection.

searchweb-scrapingbrowser-automation
96
Forks
7
Open issues
14 mo ago
Last commit
2d ago
Indexed

Overview

This is a comprehensive MCP server that provides Google search capabilities through Playwright browser automation. It features advanced anti-bot detection bypass techniques including intelligent browser fingerprint management, automatic browser state saving and restoration, and smart headless/headed mode switching. The tool can be used both as a command-line tool and as an MCP server, enabling AI assistants like Claude to perform real-time searches without relying on paid SERP APIs. The project is well-documented with TypeScript implementation, robust error handling, and comprehensive cross-platform support including Windows-specific adaptations.

Try asking AI

After installing, here are 6 things you can ask your AI assistant:

you:Enabling AI assistants to perform real-time Google searches without API keys
you:Providing a local alternative to paid SERP APIs for developers and researchers
you:Facilitating web scraping and data extraction from Google search results
you:Supporting AI applications that need current web information and real-time search capabilities
you:Does this tool violate Google's terms of service?
you:How does it bypass Google's anti-bot detection?

When to choose this

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 this

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.

Tools this server exposes

2 tools extracted from the README
  • google_searchquery: string

    Executes Google search and returns results with title, link, and snippet

  • get_htmlquery: string

    Retrieves raw HTML of Google search result pages

Comparable tools

serper-mcpsearchapi-mcpduckduckgo-searchgoogle-serp-apibing-search-mcp

Installation

# Install from source
git clone https://github.com/web-agent-master/google-search.git
cd google-search
# Install dependencies
npm install
# Compile TypeScript code
npm run build
# Link package globally (required for MCP functionality)
npm link

For Claude Desktop integration:

{
  "mcpServers": {
    "google-search": {
      "command": "npx",
      "args": ["google-search-mcp"]
    }
  }
}

FAQ

Does this tool violate Google's terms of service?
This tool is intended for learning and research purposes. Users should comply with Google's terms of service and policies, avoid sending requests too frequently, and be mindful of regional access requirements.
How does it bypass Google's anti-bot detection?
The tool uses multiple techniques including intelligent browser fingerprint management, automatic browser state saving and restoration, smart headless/headed mode switching, and randomization of device and locale settings to simulate real user behavior.

Compare google-search with

GitHub →

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