Google Custom Search MCP

by fbettag

271 downloads
Not rated
GitHub

Description

# Google Custom Search MCP Server An async Python MCP (Model Context Protocol) server that provides access to Google Custom Search API using service account authentication. ## Features - **google_search**: Perform Google Custom Search queries with configurable result count -…

About

# Google Custom Search MCP Server An async Python MCP (Model Context Protocol) server that provides access to Google Custom Search API using service account authentication. ## Features - **google_search**: Perform Google Custom Search queries with configurable result count - **Service Account Auth**: Uses Google…

Details

Author
fbettag
Downloads
271
Categories
Search, API

- Perform Google Custom Search queries with configurable result count
- Service Account authentication (file path or base64)
- Live integration tests against actual Google API
- FastMCP Python SDK integration
- Docker support with Alpine Linux image
- UV compatibility for modern Python packaging

Setting up with Highlight

This MCP is not yet compatible with Highlight’s one-click setup. However, you can still use it with Highlight by following these steps:

  1. Download and install Highlight from highlightai.com/download
  2. Navigate to the plugins tab and select "Add Custom Plugin"
  3. Configure the plugin with the settings below
    Plugin Name Google Custom Search MCP
    Command (node, npx, python, etc.)

    Please refer to the README for specific instructions on how to obtain API keys or other required environment variables.

  4. Enable "Start Automatically" if you want the plugin to start when Highlight launches

From the repository

Clone the repository, configure a Google Service Account and Custom Search Engine, then run uv run google-custom-search-mcp with the required environment variables GOOGLE_SERVICE_ACCOUNT_FILE or GOOGLE_SERVICE_ACCOUNT_BASE64 and GOOGLE_SEARCH_ENGINE_ID. The server supports STDIO mode (default) and HTTP mode (via --transport http). Docker and Helm deployments are available.

Claude Desktop / Cursor

Paste into your MCP client config file to install this server.

{
    "mcpServers": {
        "google custom search mcp": {
            "google-search": {
                "command": "uv",
                "args": [
                    "run",
                    "path/to/your/clone/of/google-custom-search-mcp"
                ],
                "env": {
                    "GOOGLE_SERVICE_ACCOUNT_FILE": "/path/to/your/service-account-key.json",
                    "GOOGLE_SEARCH_ENGINE_ID": "your-search-engine-id"
                }
            }
        }
    }
}

McpServers

{
    "google-search": {
        "command": "uv",
        "args": [
            "run",
            "path/to/your/clone/of/google-custom-search-mcp"
        ],
        "env": {
            "GOOGLE_SERVICE_ACCOUNT_FILE": "/path/to/your/service-account-key.json",
            "GOOGLE_SEARCH_ENGINE_ID": "your-search-engine-id"
        }
    }
}

Google Custom Search MCP Server

An async Python MCP (Model Context Protocol) server that provides access to Google Custom Search API using service account authentication.

Features

- google_search: Perform Google Custom Search queries with configurable result count
- Service Account Auth: Uses Google Service Account authentication (required since API keys are deprecated)
- Dual Authentication Support: File path for local dev, base64 for CI/CD
- Live Integration Tests: Real tests against Google Custom Search API
- FastMCP Integration: Uses the latest FastMCP Python SDK
- Docker Support: Alpine Linux-based Docker image for minimal footprint
- UV Compatibility: Modern Python packaging with UV

Quick Start

Prerequisites

1. Google Service Account: Create a service account in Google Cloud Console
2. Custom Search Engine ID: Create a Custom Search Engine at Programmable Search Engine
3. Enable Custom Search API: Enable the API for your Google Cloud project

Installation

```bash

No reviews yet — be the first

Sign in to leave a review

Use Google, GitHub, or an email account so ratings stay tied to real people.

Email sign in

No reviews posted yet.