Gemini MCP 服务器

Gemini MCP Server

An MCP (Model Context Protocol) server that brings Google Gemini AI capabilities to Claude Code and other MCP clients. Built following Anthropic's official MCP guidelines.

Why Gemini + Claude?

While Claude excels at many tasks, Gemini offers unique capabilities:

• Google Search Grounding - Real-time web search with source citations
• YouTube Video Analysis - Transcribe, summarize, or ask questions about videos
• Alternative Perspective - Get a second opinion from a different AI model
• Multimodal Strength - Gemini's strong video understanding capabilities

Features

Default Model: gemini-flash-latest (configurable via GEMINI_MODEL env var)

Quick Start

Prerequisites

• Node.js 18+ - Download
• Google AI API Key - Get your free key

Installation

# Clone the repository
git clone https://github.com/george7979/gemini-mcp-server.git
cd gemini-mcp-server

# Install dependencies
npm install

# Build the server
npm run build

Configuration

Option 1: Quick Install (Recommended)

Use Claude Code's built-in command:

claude mcp add gemini-mcp-server node /absolute/path/to/gemini-mcp-server/dist/index.js -e GOOGLE_API_KEY=your-api-key-here

Tip: Run pwd in the gemini-mcp-server directory to get the absolute path.

To install globally (available in all projects):

claude mcp add gemini-mcp-server node /path/to/dist/index.js -e GOOGLE_API_KEY=your-key --scope user

Option 2: Manual Configuration

Add to your Claude Code MCP settings file (~/.claude.json)

{
  "mcpServers": {
    "gemini-mcp-server": {
      "type": "stdio",
      "command": "node",
      "args": ["/absolute/path/to/gemini-mcp-server/dist/index.js"],
      "env": {
        "GOOGLE_API_KEY": "your-api-key-here",
        "GEMINI_MODEL": "gemini-flash-latest"  // optional - validated at startup
      }
    }
  }
}

Option 3: VS Code with Claude Extension

Add to .vscode/mcp.json

{
  "servers": {
    "gemini-mcp-server": {
      "type": "stdio",
      "command": "node",
      "args": ["${workspaceFolder}/path/to/gemini-mcp-server/dist/index.js"],
      "env": {
        "GOOGLE_API_KEY": "your-api-key-here",
        "GEMINI_MODEL": "gemini-flash-latest"  // optional - validated at startup
      }
    }
  }
}

Note: Replace the path with your actual installation location. You can find it with pwd in the gemini-mcp-server directory.

Verify Installation

Restart Claude Code after configuration. You should see the Gemini tools available:

gemini_generate - Generate text using Google Gemini AI
gemini_messages - Multi-turn conversation with Gemini
gemini_search   - Web search with Google Search grounding
gemini_youtube  - Analyze YouTube videos
gemini_status   - Check server status and configuration

Usage Examples

Simple Generation

Ask Gemini: "Explain quantum entanglement in simple terms"

Web Search with Citations

Search with Gemini: "What are the latest developments in AI agents in 2025?"

YouTube Analysis

Analyze this YouTube video: https://www.youtube.com/watch?v=VIDEO_ID
Prompt: "Summarize the main points in bullet format"

Multi-turn Conversation

Have a conversation with Gemini about software architecture,
maintaining context across multiple exchanges.

Tool Reference

gemini_generate

Generate text from a single prompt.

gemini_messages

Multi-turn conversation with message history.

Message format:

{
  "role": "user" | "model",
  "content": "message text"
}

gemini_search

Web search with Google Search grounding.

Returns AI response plus Search Queries and Sources with citations.

gemini_youtube

Analyze YouTube videos.

Supports both full URLs (youtube.com/watch?v=...) and short URLs (youtu.be/...).

gemini_status

Check server status and configuration.

Returns:

• active_model - Currently used model
• configured_model - Model from GEMINI_MODEL env var (if set)
• fallback_model - Default fallback model
• fallback_used - Whether fallback was triggered due to invalid model
• server_version - Server version
• api_key_configured - Whether GOOGLE_API_KEY is set

Development

# Development with hot reload
npm run dev

# Build TypeScript
npm run build

# Run compiled server
npm start

# Test with MCP Inspector
npx @modelcontextprotocol/inspector node dist/index.js

Troubleshooting

"GOOGLE_API_KEY environment variable is required"

Make sure your Claude Code configuration includes the env block with your API key.

"Invalid API key"

1. Verify your key at AI Studio
2. Make sure there are no extra spaces or quotes around the key
3. Check the key has the Gemini API enabled

"API quota exceeded"

The free tier has usage limits. Wait a few minutes and try again, or upgrade your API plan.

Tools not appearing in Claude Code

1. Verify the path in your configuration is correct (use absolute path)
2. Make sure you ran npm run build
3. Restart Claude Code after configuration changes

YouTube video errors

• Only public videos are supported (no private or unlisted)
• Video must be available in your region
• Very long videos may hit token limits (use start_offset/end_offset)

Model validation and fallback

If you configure an invalid model via GEMINI_MODEL, the server automatically falls back to gemini-flash-latest. The warning is logged to stderr but may not be visible in Claude Code.

To check your current configuration status:

1. Use the gemini_status tool - it shows active model and whether fallback occurred
2. Run Claude Code with --verbose flag to see MCP server logs

Project Structure

gemini-mcp-server/
├── src/
│   └── index.ts          # Server implementation
├── dist/                 # Compiled output
├── docs/
│   ├── PRD.md            # Product requirements
│   ├── PLAN.md           # Implementation roadmap
│   └── TECH.md           # Technical specification
├── package.json
├── tsconfig.json
├── .env.example          # Environment template
├── README.md             # This file
└── CLAUDE.md             # AI assistant context

Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

License

MIT License - see LICENSE for details.

Acknowledgments

• Anthropic - MCP Protocol and Claude
• Google - Gemini AI API
• Model Context Protocol - Protocol specification

Made with Claude Code following Anthropic's MCP guidelines