Google Custom Search

🚀 MCP Server Using Google Custom Search

An MCP (Model Context Protocol) server that provides web search functionality, leveraging the Google Custom Search API and web content extraction capabilities.

🚀 Quick Start

Obtain Google API Key and Search Engine ID

1. Create a Google Cloud project:

   • Visit the Google Cloud Console.
   • Create a new project or select an existing one.
   • Enable project billing.

2. Enable the Custom Search API:

   • Visit the API Library.
   • Search for "Custom Search API".
   • Click "Enable".

3. Get an API key:

   • Visit Credentials.
   • Click "Create credentials" > "API key".
   • Copy your API key.
   • (Optional) Restrict the key to only be used for the Custom Search API.

4. Create a custom search engine:

   • Visit the Programmable Search Engine.
   • Enter the sites you want to search (use www.google.com for a full - web search).
   • Click "Create".
   • In the next step, click "Customize".
   • Enable "Search the entire web" in the settings.
   • Copy your search engine ID (cx).

📦 Installation

Install via Smithery

To automatically install the Google Custom Search server for Claude Desktop via Smithery:

npx -y @smithery/cli install @adenot/mcp-google-search --client claude

When using it with Claude Desktop, add the server configuration containing your Google API credentials:

On MacOS: ~/Library/Application Support/Claude/claude_desktop_config.json On Windows: %APPDATA%/Claude/claude_desktop_config.json

{
  "mcpServers": {
    "google-search": {
      "command": "npx",
      "args": [
        "-y",
        "@adenot/mcp-google-search"
      ],
      "env": {
        "GOOGLE_API_KEY": "your-api-key-here",
        "GOOGLE_SEARCH_ENGINE_ID": "your-search-engine-id-here"
      }
    }
  }
}

💻 Usage Examples

Search Tool

{
  "name": "search",
  "arguments": {
    "query": "Your search query",
    "num": 5 // Optional, default is 5, range 1 - 10
  }
}

Web Reader Tool

{
  "name": "extract",
  "arguments": {
    "url": "Web page link"
  }
}

Example Responses

Search Tool:

{
  "items": [
    {
      "title": "Result Title 1",
      "link": "https://example.com/1",
      "snippet": "Result Excerpt 1"
    },
    // Other results...
  ]
}

Web Reader Tool:

{
  "title": "Page Title",
  "text": "Page body content..."
}

🛠️ Development

Install dependencies:

npm install

Build the server:

npm run build

Automatically rebuild during development:

npm run watch

✨ Features

Search Tool

Perform web searches using the Google Custom Search API:

• Search the entire web or specific sites.
• Control the number of results (1 - 10).
• Get structured results, including titles, links, and excerpts.

Web Reader Tool

Extract content from any web page:

• Crawl and parse web page content.
• Extract the page title and body text.
• Clean the content to remove scripts and styles.
• Return structured data, including title, text, and URL.

⚙️ Debugging

Use the DEBUG=mcp-google-search environment variable to enable debug output:

DEBUG=mcp-google-search npm start

⚠️ Important Note

• Ensure you have permission to search the target sites.
• Comply with Google API's terms of service and usage restrictions.