Gemini MCP Server

English Version

Claude Code Gemini AI MCP .

1. ** **: Gemini

2. ** **: Gemini CLI Claude Code Claude Code

3. ** **: Claude Code

   gemini CLI .

• Python 3.8+
• uv
• Gemini CLI CLI

1.

git clone https://github.com/InfolabAI/gemini-cli-mcp.git
cd gemini-cli-mcp

• uv

uv (Python ):

# Linux/macOS ()
curl -LsSf https://astral.sh/uv/install.sh | sh

# macOS Homebrew
brew install uv

# Windows PowerShell
powershell -c "irm https://astral.sh/uv/install.ps1 | iex"

#  
uv --version

• :

uv sync

Tools

• run_gemini
  • Gemini , , URL
  • :
    • prompt (string): Gemini
    • file_dir_url_path (string): , URL
    • working_directory (string): ()

Configuration

Claude Code MCP :

Linux , ~/.claude.json :

{
  "mcpServers": {
    "gemini": {
      "command": "uv",
      "args": [
        "--directory",
        "/path/to/your/project",
        "run",
        "python",
        "/path/to/your/project/gemini_mcp_server.py"
      ],
      "env": {}
    }
  }
}

****: /path/to/your/project/ git .

• ( , )

uv run python gemini_mcp_server.py

uv run python test_gemini_mcp.py

• Debugging

MCP stdio . MCP Inspector .

subprocess.run shell=True . MCP . , Gemini CLI , . , MCP stdio stdin/stdout , CLI ( Node.js ) tty . run_gemini.py tty shell=True .

Claude Code MCP Gemini CLI . , :

• Claude Code Gemini Gemini

  Welcome to Claude Code!                         



> gemini  gemini_mcp_server.py   

 I'll analyze the gemini_mcp_server.py file using the Gemini MCP server to provide a summary.

 gemini-mcp-server:run_gemini (MCP)(prompt: " Python      . , ,    .", file_dir_url_path:
                                    "gemini_mcp_server.py")
   { 
    "result": "[dotenv@16.6.0] injecting env (2) from ../../../.env\n[dotenv@16.6.0] injecting env (2) from ../../../.env\n[dotenv@16.6.0] injecting env (2) from ../../../.env\nMCP STDERR (github): GitHub MCP Server running on stdio\n\nMCP STDERR (sequential-thinking): Sequential Thinking MCP Server running on stdio\n\nMCP STDERR (context7): Context7 Documentation MCP Server running on stdio\n\n Python (`gemini_mcp_server.py`) Gemini CLI  MCP(Multi-Agent Communication Protocol)         .\n\n**   :**\n\n*   **FastMCP  **: `FastMCP(\"Gemini MCP Server\")`  `mcp` MCP   .    MCP         .\n*   **`run_gemini`  ()**:\n    *   `@mcp.tool()`   ,   MCP       \"\"  .\n    *   ****: `prompt` (Gemini  ) `file_dir_url_path` ( ,   URL ) .\n    *   ** **:\n 1.  `file_dir_url_path` `prompt`  Gemini CLI    .\n        2.  `subprocess.run`   `gemini` CLI  .   `shell=True`      ,  MCP      .\n        3.  `gemini -m gemini-2.5-flash -p \"...\"`   .\n        4.        `subprocess.TimeoutExpired` (60  ), `FileNotFoundError` (Gemini CLI   ),   `Exception`      .\n        5.  Gemini CLI    (`stdout`)  ,     .\n*   ** **: `if __name__ == \"__main__\":`   `mcp.run()`  MCP  .         .\n\n**:**\n\n`gemini_mcp_server.py` `FastMCP`   Gemini CLI    (`run_gemini` )   .     //URL    Gemini CLI    ,        ."
    }

• Gemini Claude Code

 gemini_mcp_server.py Gemini CLI MCP   . FastMCP   run_gemini  ,     Gemini CLI   . 60     .

MIT .