MCP Google Custom Search Server

Provides MCP-compliant web search using Google Custom Search API for language models.

typescript

GitHub Stars

typescript

Language

6 months ago

First Indexed

2 months ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

{
  "mcpServers": {
    "mcp-mirror-alexatnordnet_mcp-google-custom-search-server": {
      "command": "node",
      "args": [
        "/absolute/path/to/mcp-google-custom-search-server/build/index.js"
      ],
      "env": {
        "GOOGLE_API_KEY": "your-api-key",
        "GOOGLE_SEARCH_ENGINE_ID": "your-search-engine-id"
      }
    }
  }
}

You can extend your language models with web search capabilities by running an MCP-compliant server that talks to Google Custom Search. This server provides structured, formatted search results to models via a standard interface, enabling practical, real-time web queries within your applications.

How to use

You access this MCP server through an MCP client. Start the local server, then use an MCP tool named search to perform web queries. The server returns formatted results with titles, URLs, and descriptions, suitable for ingestion by your model. You can configure the number of results per query (up to 10) and rely on input validation to ensure queries are well-formed.

How to install

Prerequisites: you need Node.js v18 or higher and npm. You also require a Google Cloud project with the Custom Search API enabled and a configured Custom Search Engine ID.

Step 1: Clone the project directory and navigate into it.

Step 2: Install dependencies.

Step 3: Create an environment file with your Google API key and Search Engine ID.

Step 4: Build the server.

Step 5: Start the MCP server.

Configuration and usage notes

Environment variables you must provide for the server to run are your Google API key and your Custom Search Engine ID. These values are passed to the server at runtime and used to authenticate requests to Google.

Claude Desktop integration: you can register the MCP server in Claude Desktop so you can invoke web searches directly from Claude. The config below shows how to wire the local server into Claude Desktop. Replace "/absolute/path/to/mcp-google-custom-search-server/build/index.js" with the actual path to your built server.

{
  "mcpServers": {
    "google-search": {
      "command": "node",
      "args": [
        "/absolute/path/to/mcp-google-custom-search-server/build/index.js"
      ],
      "env": {
        "GOOGLE_API_KEY": "your-api-key",
        "GOOGLE_SEARCH_ENGINE_ID": "your-search-engine-id"
      }
    }
  }
}

Available tools

search

Performs a web search using Google Custom Search API and returns formatted results including titles, URLs, and descriptions.

Built by

VeilStrat

AI signals for GTM teams