This project provides a Model Context Protocol (MCP) server that integrates the Google Gemini CLI with AI assistants like Claude Code. It enables AI models to leverage Gemini's large token window for in-depth analysis of large files and codebases, improving efficiency and reducing token consumption for complex tasks.

How It Works

The tool acts as a bridge, translating AI assistant commands into calls to the Gemini CLI. It utilizes the MCP protocol to manage context and communication between the AI and the Gemini CLI. The @ syntax allows users to reference files or directories directly within prompts, enabling Gemini to process them efficiently. The inclusion of a sandbox mode offers a secure environment for executing code or scripts.

Quick Start & Requirements

• Install/Run: npx -y gemini-mcp-tool
• Prerequisites: Node.js (v16.0.0 or higher), Google Gemini CLI installed and configured.
• Documentation: Wiki documentation is available.

Highlighted Details

• Enables AI assistants to interact with Google Gemini CLI.
• Leverages Gemini's large token window for analyzing large files and codebases.
• Supports file/directory referencing via the @ syntax.
• Includes a sandbox mode for safe code execution.

Maintenance & Community

The project is open source with an active npm presence. Contribution guidelines are available.

Licensing & Compatibility

Licensed under the MIT License, permitting commercial use and integration with closed-source projects.

Limitations & Caveats

This is an unofficial, third-party tool and is not affiliated with, endorsed, or sponsored by Google. Potential conflicts may arise if other MCP tools are installed and write to stdout. Nested prompts are possible but do not automatically fall back on quota exhaustion.