infolabai/gemini-cli-mcp

Developer tools

This page was translated automatically. Technical details may differ from the original. View the original

30 stars 1 forks Quality 56 Trend 11MIT

Repository Install

It is a tool that allows Gemini AI to be used as an MCP server in Claude Code.

Overview

It is a tool that allows Gemini AI to be used as an MCP server in Claude Code. 1. : Utilize Gemini's massive context window to analyze large files and directories at once 2. : Save on Claude Code token usage while leveraging Claude Code's powerful features using the free Gemini CLI 3. : Easily integrate into existing Claude Code workflows This server operates using locally installed gemini CLI tools. - Python 3.8+ - uv Package Manager - Gemini CLI CLI Tool UV must be installed at the system level (operating independently of the Python virtual environment): - - Handles large amounts of information for files, directories, and URLs using Gemini - Input: - prompt (string): Prompt to send to Gemini - file_dir_url_path (string): The path to the file, directory, or URL to analyze - working_directory (string): Task directory (required) For Linux, add the following settings to the ~/.claude.json file: : Change /path/to/your/project/ to the path of the downloaded git project.

Overview (machine translated)

English Version

It is a tool that allows Gemini AI to be used as an MCP server in Claude Code.

Key Benefits

Large File Analysis: Utilize Gemini’s massive context window to analyze large files and directories at once
Token Savings: Use the free Gemini CLI to save on Claude Code’s token usage while leveraging its powerful unique features
Easy Integration: Easily integrated into existing Claude Code workflows

This server operates using the locally installed gemini CLI tool.

Prerequisites

Python 3.8+
UV Package Manager
Gemini CLI CLI Tool

Quick Start

1. Storage clone

UV installation

UV must be installed at the system level (operating independently of the Python virtual environment):

Installation of dependencies:

Tools

run_gemini
Handle large amounts of information from files, directories, and URLs using Gemini
Input:
prompt (string): Prompt to send to Gemini
file_dir_url_path (string): The path to the file, directory, or URL to analyze
working_directory (string): Task directory (required)

Configuration

To use with Claude Code, add the MCP server settings:

For Linux, add the following settings to the ~/.claude.json file:

Note: Change /path/to/your/project/ to the path of the downloaded git project.

Development

Run directly on the server (after running, it should be stopped)
Testing
Debugging

Since the MCP server communicates via stdio, debugging can be difficult. We recommend using MCP Inspector.

Points to note

When using subprocess.run, you must use the shell=True option. This is to execute commands through the shell in the MCP server environment. Without this option, you may enter a waiting state when executing Gemini CLI commands.

This page was translated automatically. Technical details may differ from the original.

Install

This server does not publish a one-line install command.

Open the repository installation guide New to MCP? Start with the mcpskill.org guide →

Configuration

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