Claude Gemini MCP Integration
๐ What's Coming Next: I'm building an AI Agent Feedback Loop System that enables intelligent collaboration between Claude Code and Gemini AI. This will create a continuous improvement cycle where both AI agents learn from each other's suggestions, creating smarter code analysis and more contextual development assistance. Starting with Claude Code, then expanding to other IDEs. Stay tuned!
A lightweight integration that brings Google's Gemini AI capabilities to Claude Code through MCP (Model Context Protocol)
This project connects Claude Code (your coding assistant) with Google's Gemini AI models. Think of it as adding a second AI expert to your development team - one that can read and understand massive amounts of code at once (1M+ tokens, which is like reading hundreds of code files simultaneously).
With this integration, you can ask Gemini questions about your code, get security reviews, performance suggestions, and architectural advice - all without leaving your coding environment. It automatically chooses the right AI model for each task: fast responses for quick questions, deeper analysis for complex problems.
Table of Contents
- Key Features
- How It Works
- Architecture Overview
- Quick Start
- Usage Examples
- Benefits
- Usage Examples with Slash Commands
- What's Next?
- Need Help?
- Further Documentation
- Changelog
- Contributing
- License
- Credits
Key Features
- Quick Query - Ask Gemini any development question instantly
- Code Analysis - Deep analysis with security, performance, and architecture insights
- Codebase Analysis - Full project analysis using Gemini's massive context window
- Automated Hooks - Pre-edit analysis, pre-commit review, and session summaries
- 20+ Slash Commands - Simple shortcuts like
/g,/analyze,/security - Smart Model Selection - Flash for speed, Pro for depth, automatic fallback
- Real-time Streaming - Live output with progress indicators
How It Works
Claude Code โโ MCP Server โโ Gemini CLI/API โโ Google Gemini Models
โ
Smart Model Selection
(Flash for speed, Pro for depth)
Architecture Overview
The Gemini MCP server uses a shared architecture where one installation serves multiple AI clients and projects:
Claude Desktop โ Claude Code โ Cursor IDE โ VS Code + Extensions
โ โ โ
โโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโ
โ
โโโโโโโโโโโโโโโโโโโโโโโโโโโ
โ MCP Protocol โ
โ (Tool Requests) โ
โโโโโโโโโโโโโโโโโโโโโโโโโโโ
โ
โโโโโโโโโโโโโโโโโโโ
โ Gemini MCP โ
โ Server โ
โ (Python/Shell) โ
โโโโโโโโโโโโโโโโโโโ
โ
โโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโ
โ โ โ
โโโโโโโโโโโโโโโโโโโ โโโโโโโโโโโโโโโโโโโ โ
โ Gemini API โ โ Gemini CLI โ โ
โ (Direct HTTP) โ โ (Shell Command)โ โ
โโโโโโโโโโโโโโโโโโโ โโโโโโโโโโโโโโโโโโโ โ
โ โ โ
โโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโ
โ
โโโโโโโโโโโโโโโโโโโโโโโโโโโ
โ Google Gemini AI โ
โ (1M+ Token Context) โ
โโโโโโโโโโโโโโโโโโโโโโโโโโโ
Key Benefits
- One installation serves all AI clients and projects
- No project pollution - keeps your projects clean
- Easy maintenance - update once, benefits everywhere
- Smart fallbacks - API-first approach with CLI backup
File Structure
~/mcp-servers/ โ Central location for all MCP servers
โโโ shared-mcp-env/ โ Shared virtual environment
โ โโโ bin/python โ Python interpreter for all MCPs
โ โโโ lib/python3.x/site-packages/ โ Shared dependencies (mcp, google-generativeai, etc.)
โโโ gemini-mcp/ โ Complete Gemini MCP package
โโโ gemini_mcp_server.py โ Main MCP server
โโโ .claude/ โ Complete slash commands system
โโโ hooks.json โ Hook definitions
โโโ commands/ โ Native slash commands (10+ commands)
โ โโโ gemini.md โ /gemini command
โ โโโ analyze.md โ /analyze command
โ โโโ ... โ Other command definitions
โโโ scripts/
โโโ slim_gemini_hook.py โ Hook execution script
Project Structure (with hooks enabled)
your-project/
โโโ .claude โ ~/mcp-servers/gemini-mcp/.claude โ Symlink to shared hooks
โโโ src/ โ Your project files
โโโ README.md
โโโ (no venv or MCP files needed!) โ Clean project structure
Multi-Client Support
The shared MCP architecture supports multiple AI clients simultaneously:
Supported Clients:
- Claude Desktop - Core MCP tools only
- Claude Code - Core MCP tools + hooks (if configured)
- VS Code with Claude Code extension - Core MCP tools + hooks (if configured)
- Cursor IDE - Core MCP tools only
- Windsurf - Core MCP tools only
- VS Code with other MCP extensions - Core MCP tools only
- Any MCP-compatible client - Core MCP tools only
Important: Hook functionality (.claude/hooks.json) is exclusive to Claude Code ecosystem (Claude Code standalone + VS Code with Claude Code extension). No other AI client currently supports this automation system.
Quick Start
New to this project? Here's what you need to do:
- Get your Google API key from Google AI Studio
- Follow the complete setup guide in docs/SETUP/SETUP.md
- Test the integration with a simple query
- Explore the 20+ slash commands in docs/README-SLASH-COMMANDS.md
Installation time: ~5 minutes | Prerequisites: Python 3.10+, Node.js 16+
Usage Examples
Quick Development Questions
Use gemini_quick_query for:
- "How do I implement JWT authentication in Node.js?"
- "What's the difference between useEffect and useLayoutEffect?"
- "Best practices for error handling in Python async functions"
Code Analysis
Use gemini_analyze_code for:
- Security review of authentication functions
- Performance analysis of database queries
- Architecture review before major refactoring
Full Project Analysis
Use gemini_codebase_analysis for:
- Overall architecture assessment
- Security vulnerability scanning
- Performance bottleneck identification
Automated Hooks System
The hooks system provides intelligent automation that runs at key development moments:
Pre-edit Analysis:
- Automatically analyzes files before Claude Code edits them
- Provides context about security, performance, and architecture concerns
- Helps prevent issues by informing Claude Code before changes are made
Pre-commit Review:
- Analyzes staged changes before git commits
- Reviews code for critical bugs, security vulnerabilities, and quality issues
- Provides final quality check before code enters version control
Session Summary:
- Generates brief recap when Claude Code session ends
- Highlights key changes made and potential next steps
- Maintains development context between sessions
Benefits
For Developers
- Instant access to Gemini's advanced AI capabilities
- Seamless integration within Claude Code workflow
- Smart model selection - fast responses when speed matters, deep analysis when needed
- Real-time feedback during long analysis tasks
For Code Quality
- Security analysis using Gemini's latest training data
- Performance insights from large-scale pattern recognition
- Architecture guidance based on current best practices
- Error prevention through pre-edit analysis
For Productivity
- Reduced context switching - stay in Claude Code
- Faster debugging with AI-powered error analysis
- Better decisions through comprehensive code review
- Learning acceleration via instant expert guidance
Usage Examples with Slash Commands
The slash commands provide a much simpler way to access Gemini's powerful analysis without remembering the full MCP