Serper MCP Server

Author:@garymengcom
View on GitHub

Overview

Serper MCP Server is a Model Context Protocol server. It provides Google Search functionality through the Serper service, allowing Large Language Models (LLMs) to access and retrieve search result information from Google.

To use this server, install it via package managers like uv or pip. Then, configure your MCP client to include this server. You must also set your Serper API key in your environment variables for authentication.

Its key features include delivering Google Search results for LLMs, supporting various search parameters such as location, language, and time period, and offering easy integration with MCP clients using standard installation tools.

Common use cases are enabling LLMs to perform real-time Google searches, integrating search capabilities into AI applications, and assisting research by providing current information from the web.

Frequently asked questions include installation instructions via uv or pip, the requirement of a valid Serper API key and a properly configured MCP client, and the availability of the MCP inspector as a debugging tool for the server.

Configuration

{
  "mcpServers": {
    "serper": {
      "command": "uvx",
      "args": [
        "serper-mcp-server"
      ],
      "env": {
        "SERPER_API_KEY": "<Your Serper API key>"
      }
    }
  }
}