Google Custom Search

A Model Context Protocol server that provides search capabilities using a Google CSE (custom search engine).

python

GitHub Stars

python

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": {
    "richard-weiss-mcp-google-cse": {
      "command": "uvx",
      "args": [
        "mcp-google-cse"
      ],
      "env": {
        "API_KEY": "YOUR_API_KEY",
        "ENGINE_ID": "YOUR_ENGINE_ID",
        "RESULT_NUM": "10",
        "GEOLOCATION": "us",
        "COUNTRY_REGION": "us",
        "RESULT_LANGUAGE": "lang_en"
      }
    }
  }
}

You can run a Google Custom Search Engine MCP Server that exposes a search capability to your MCP client. It lets you perform Google-like searches through a custom search engine and return the results (title, link, snippet) to your agent, which can then fetch content from those results with other tools. This is useful for adding live web search to your toolchain while keeping content retrieval separate.

How to use

Use this MCP server with your MCP client to perform Google-style searches via a custom search engine. The server exposes a single tool named google_search that accepts a search term and returns a list of results. You can combine this with other MCP servers (for example, to fetch the actual page content from the results) to build a richer browsing workflow.

Typical usage patterns include: 1) Send a search query to obtain a set of relevant results. 2) Pass those results to a content-fetching server to retrieve full documents. 3) Chain results with other tools to perform deeper analysis or extract data from the pages.

How to install

Prerequisites: You need Python installed to run the pip installation option, or you can use uvx to run a prebuilt MCP server binary. Choose one of the following installation methods.

Option A — Install via pip and run with Python

pip install mcp-google-cse

python -m mcp-google-cse

Option B — Run via uvx (recommended for quick starts)

uvx

Then run the specific MCP module with the uvx-based command configuration (see configuration section for exact env vars).

## Additional content

Configuration notes and environment variables are essential for getting the Google Custom Search Engine MCP Server running. You provide credentials for the custom search engine and optional search parameters to tailor results.

Important quotas: The free plan allows 100 searches per day. Each tool call counts as one search. If you expect higher usage, set up billing or use an alternative server.

Example configuration blocks show how to wire up the service in your Claude Desktop config. You will supply `API_KEY` and `ENGINE_ID` along with optional regional and language preferences.

## Available tools

### google\_search

Performs a search on the configured custom search engine and returns a list of results with title, link, and snippet for each result.

Built by

VeilStrat

AI signals for GTM teams