Google Search MCP

v1.0.0Search & Data Extractionstable

A Playwright-based Node.js tool that bypasses search engine anti-scraping mechanisms to execute Google searches. Local alternative to SERP APIs with MCP server integration.

aigoogle-searchllmmcp-serverweb-scraping
Share:
596
Stars
0
Downloads
0
Weekly
0/5

What is Google Search MCP?

Google Search MCP is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to playwright-based node.js tool that bypasses search engine anti-scraping mechanisms to execute google searches. local alternative to serp apis with mcp server integration.

A Playwright-based Node.js tool that bypasses search engine anti-scraping mechanisms to execute Google searches. Local alternative to SERP APIs with MCP server integration.

This server falls under the Search & Data Extraction category on MCPgee, the world's largest MCP server directory with 33,000+ servers.

Features

  • search

Use Cases

Bypass Google anti-scraping with Playwright
Local SERP alternative
Real-time search results
LicenseMIT
Languagetypescript
Versionv1.0.0
UpdatedMay 19, 2026
Statushealthy
Maintenanceactive

Works with

ClaudeOpenAIwindowsmacoslinux

Installation

NPM

npx -y google-search

Manual Installation

npx -y google-search

Configuration

Configuration Details

Config File

claude_desktop_config.json

Performance

Response Metrics

Response Time< 200ms
ThroughputMedium

Resource Usage

Memory UsageLow
CPU UsageLow

How to Set Up and Use Google Search MCP

Google Search MCP is a Playwright-based Node.js tool that performs real Google searches from the command line or as an MCP server, bypassing anti-scraping mechanisms without requiring any API key or paid SERP service. It returns structured JSON results (title, URL, snippet) or raw HTML and integrates directly with Claude Desktop and other MCP clients to give AI assistants live web search capabilities. It is a practical local alternative to paid search APIs for developers who need real-time results in their AI workflows.

Prerequisites

  • Node.js 18 or higher installed
  • npm or npx available on your PATH
  • Google Chrome or Chromium browser available (Playwright will manage a separate instance)
  • An MCP client such as Claude Desktop or Claude Code CLI
  • No API keys required
1

Clone and build from source

Clone the repository, install dependencies, build the TypeScript source, and link the binary globally so it is accessible as a command.

git clone https://github.com/web-agent-master/google-search.git
cd google-search
npm install
npm run build
npm link
2

Test a search from the command line

Verify the tool works by running a quick Google search. Results are returned as JSON with title, link, and snippet for each result.

google-search "latest AI research papers"
# Limit results and get raw HTML:
google-search "openai news" --limit 5 --get-html --save-html
3

Add to your MCP client configuration

Edit your claude_desktop_config.json to register the Google Search server. The command uses the npx invocation to run the built package.

{
  "mcpServers": {
    "google-search": {
      "command": "npx",
      "args": ["google-search-mcp"]
    }
  }
}
4

Restart your MCP client

Restart Claude Desktop or your chosen client to load the new MCP server. The search tool should appear in the available tools.

5

Test with a natural language search prompt

Ask Claude to search Google for something. The server will use Playwright to perform the search and return structured results.

Google Search MCP Examples

Client configuration

Register the Google Search MCP server in your claude_desktop_config.json file.

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

Prompts to try

Use these prompts once Google Search MCP is connected to your AI assistant.

- "Search Google for the latest news about Claude AI and summarize the top 5 results"
- "Find the official documentation page for the Playwright Node.js library"
- "Search for 'TypeScript MCP server tutorial' and list the titles and URLs"
- "Look up current pricing for AWS Lambda and compare with Google Cloud Functions"

Troubleshooting Google Search MCP

Search returns empty results or the browser fails to launch

Run with the --no-headless flag to see the browser window and diagnose what is happening: google-search "test" --no-headless. If Playwright cannot find a browser, run 'npx playwright install chromium' to download it.

Google blocks the search with a CAPTCHA or bot detection page

The tool uses browser fingerprinting and state restoration to avoid detection, but may still be blocked under heavy use. Use the --state-file flag to persist browser state between runs, which helps maintain a trusted browser session: google-search "query" --state-file ~/.google-search-state.json

MCP server command not found after npm link

Run 'npm list -g | grep google-search' to confirm the global link. If not found, run 'npm link' again from inside the cloned repository directory. On macOS/Linux you may need sudo for global npm links.

Frequently Asked Questions about Google Search MCP

What is Google Search MCP?

Google Search MCP is a Model Context Protocol (MCP) server that playwright-based node.js tool that bypasses search engine anti-scraping mechanisms to execute google searches. local alternative to serp apis with mcp server integration. It connects AI assistants to external tools and data sources through a standardized interface.

How do I install Google Search MCP?

Install via npm with the command: npx -y google-search. Then add the server configuration to your AI client's JSON config file (e.g., claude_desktop_config.json or .cursor/mcp.json).

Which AI clients work with Google Search MCP?

Google Search MCP works with all major MCP-compatible AI clients including Claude Desktop, Claude Code, Cursor, VS Code (GitHub Copilot), Windsurf, and Cline.

Is Google Search MCP free to use?

Yes, Google Search MCP is open source and available under the MIT license. You can use it freely in both personal and commercial projects.

Browse More Search & Data Extraction MCP Servers

Explore all search & data extraction servers available in the MCPgee directory. Each server includes setup guides for Claude, Cursor, and VS Code.

Quick Config Preview

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

Add this to your claude_desktop_config.json or .cursor/mcp.json

Read the full setup guide →

Ready to use Google Search MCP?

Browse our complete directory of 33,000+ MCP servers, read setup guides for your editor, and start building with the Model Context Protocol.

33,000+ ServersFree & Open SourceStep-by-Step Guides