Google Gemini

Model Context Protocol (MCP) server for Gemimi integration, built on FastMCP.

This server is implemented in Python, with fastmcp.

Quick Start

1. Build the Docker image:

docker build -t gemini-mcp-server .

Integration with Cursor/Claude

In MCP Settings -> Add MCP server, add this config:

{
  "mcpServers": {
    "gemini": {
      "command": "docker",
      "args": [
        "run",
        "--rm",
        "-i",
        "--network",
        "host",
        "-e",
        "GEMINI_API_KEY",
        "-e",
        "GEMINI_MODEL",
        "-e",
        "GEMINI_BASE_URL",
        "-e",
        "HTTP_PROXY",
        "-e",
        "HTTPS_PROXY",
        "gemini-mcp-server:latest"
      ],
      "env": {
        "GEMINI_API_KEY":"your_api_key_here",
        "GEMINI_MODEL":"gemini-2.5-flash",
        "GEMINI_BASE_URL":"https://generativelanguage.googleapis.com/v1beta/openai/",
        "HTTP_PROXY":"http://127.0.0.1:17890",
        "HTTPS_PROXY":"http://127.0.0.1:17890"

      }
    }
  }
}

Note: Don't forget to replace GEMINI_API_KEY、GEMINI_MODEL、GEMINI_BASE_URL、HTTP_PROXY、HTTPS_PROXY values with your actual Gemimi credentials and instance URL.