MCP Google Custom Search Server

A Model Context Protocol (MCP) server that provides web search capabilities through Google's Custom Search API. This server enables Language Learning Models (LLMs) to perform web searches using a standardized interface.

🌟 Features

• Seamless integration with Google Custom Search API
• Model Context Protocol (MCP) compliant server implementation
• Type-safe implementation using TypeScript
• Environment variable configuration
• Input validation using Zod
• Configurable search results (up to 10 per query)
• Formatted search results including titles, URLs, and descriptions
• Error handling and validation
• Compatible with Claude Desktop and other MCP clients

📋 Prerequisites

Before you begin, ensure you have:

1. A Google Cloud Project with Custom Search API enabled
   • Visit Google Cloud Console
   • Enable the Custom Search API
   • Create API credentials
2. A Custom Search Engine ID
   • Visit Programmable Search Engine
   • Create a new search engine
   • Get your Search Engine ID
3. Local development requirements:
   • Node.js (v18 or higher)
   • npm (comes with Node.js)

🚀 Quick Start

1. Clone the repository:
   git clone https://github.com/yourusername/mcp-google-custom-search-server.git cd mcp-google-custom-search-server
2. Install dependencies:
   npm install
3. Create a .env file:
   GOOGLE_API_KEY=your-api-key GOOGLE_SEARCH_ENGINE_ID=your-search-engine-id
4. Build the server:
   npm run build
5. Start the server:
   npm start

🔧 Configuration

Environment Variables

Claude Desktop Integration

Add this configuration to your Claude Desktop config file (typically located at ~/Library/Application Support/Claude/claude_desktop_config.json):

📖 API Reference

Available Tools

search

Performs a web search using Google Custom Search API.

Parameters:

• query (string, required): The search query to execute
• numResults (number, optional): Number of results to return
  • Default: 5
  • Maximum: 10

Example Response:

🛠️ Development

Project Structure

Available Scripts

• npm run build: Compile TypeScript to JavaScript
• npm start: Start the MCP server
• npm run dev: Watch mode for development

Testing

1. Using MCP Inspector:
   npx @modelcontextprotocol/inspector node build/index.js
2. Manual testing with example queries:
   # After starting the server {"jsonrpc":"2.0","id":1,"method":"callTool","params":{"name":"search","arguments":{"query":"example search"}}}

📝 License

This project is licensed under the MIT License - see the LICENSE file for details.

🙏 Acknowledgments

• Built with Model Context Protocol (MCP)
• Uses Google's Custom Search API
• Inspired by the need for better search capabilities in LLM applications