9.5 KiB
9.5 KiB
Gemini MCP Server Documentation
Welcome to the comprehensive documentation for the Gemini MCP Server - a sophisticated Model Context Protocol server that enables Claude to access Google's Gemini AI models through specialized tools for AI-assisted development workflows.
📖 Documentation Overview
This documentation is organized into four main categories to serve different audiences and use cases:
🚀 For End Users
- Installation Guide - Set up the server locally or with Docker
- Configuration - Configure the server for your environment
- Troubleshooting - Common issues and solutions
🛠️ For Developers
- Development Setup - Set up your development environment
- Development Workflows - Git workflows, testing, and collaboration patterns
- Code Style Guide - Coding standards and best practices
- Testing Strategy - Testing approaches and quality assurance
- Repository Overview - Understanding the codebase structure
🏗️ For System Architects
- Architecture Overview - High-level system design and components
- Component Details - Detailed component descriptions and interactions
- Data Flow Patterns - How data moves through the system
- Architecture Decisions - Architecture Decision Records (ADRs)
🔧 For API Users
- MCP Protocol - Model Context Protocol implementation details
- Tool Reference - Individual tool API documentation
🎯 Quick Start Paths
New User Journey
- Install the Server → Get up and running quickly
- Configure Your Setup → Customize for your environment
- Start Using Tools → Explore AI-assisted workflows
- Troubleshoot Issues → Resolve common problems
Developer Journey
- Set Up Development → Prepare your dev environment
- Understand the Codebase → Navigate the repository
- Follow Workflows → Git, testing, and collaboration
- Code Quality Standards → Maintain code quality
Architect Journey
- System Overview → Understand the high-level design
- Component Architecture → Deep dive into system parts
- Data Flow Analysis → Trace information flow
- Decision Context → Understand design choices
🛠️ Tool Reference
The server provides six specialized tools for different AI collaboration scenarios:
| Tool | Purpose | Best For | Documentation |
|---|---|---|---|
| chat | Quick questions, brainstorming | Immediate answers, idea exploration | Low complexity, fast iteration |
| thinkdeep | Complex analysis, strategic planning | Architecture decisions, system design | High complexity, deep analysis |
| analyze | Code exploration, system understanding | Codebase comprehension, dependency analysis | Medium complexity, systematic exploration |
| codereview | Code quality, security, bug detection | PR reviews, security audits | Quality assurance, comprehensive validation |
| debug | Root cause analysis, error investigation | Bug fixing, performance issues | Problem-solving, systematic debugging |
| precommit | Automated quality gates | Pre-commit validation, change analysis | Quality gates, automated validation |
Tool Selection Guide
For Quick Tasks: Start with chat for immediate answers and brainstorming
For Complex Planning: Use thinkdeep for architecture and strategic decisions
For Code Understanding: Use analyze to explore and understand existing code
For Quality Assurance: Use codereview and precommit for validation
For Problem Solving: Use debug for systematic error investigation
🔄 Collaboration Framework
This project follows the CLAUDE.md Collaboration Framework which defines:
- Tool Selection Matrix: Guidelines for choosing the right tool for each task
- Memory Bank Integration: Context preservation across development sessions
- Quality Gates: Mandatory validation and review processes
- Documentation Standards: Comprehensive documentation requirements
Key Collaboration Patterns
- Complex Tasks (>3 steps): Always use TodoWrite to plan and track progress
- Architecture Decisions: Must involve
thinkdeepfor exploration before implementation - Code Reviews: All significant changes require
codereviewanalysis before committing - Documentation Updates: Any code change must include corresponding documentation updates
📚 Additional Resources
Configuration Examples
- macOS Setup - Local development on macOS
- WSL Setup - Windows Subsystem for Linux
- Docker Setup - Container-based deployment
Project Information
- Main README - Project overview and quick start
- Contributing Guidelines - How to contribute to the project
- License - MIT License details
- Collaboration Framework - Development collaboration patterns
Memory Bank System
The project uses a Memory Bank system for context preservation:
- Product Context - Project goals and architecture
- Active Context - Current development status
- Decision Log - Architectural decisions and rationale
- Progress Tracking - Task completion and milestones
🎨 Documentation Standards
For Technical Audiences
- Code Context: All explanations include specific file and line number references (
file_path:line_number) - Architecture Focus: Explain why decisions were made, not just what was implemented
- Data Flow: Trace data through the system with concrete examples
- Error Scenarios: Document failure modes and recovery strategies
For Non-Technical Audiences
- Plain Language: Avoid jargon, explain technical terms when necessary
- Purpose-Driven: Start with "what problem does this solve?"
- Visual Aids: Use diagrams and flowcharts where helpful
- Practical Examples: Show real usage scenarios
🔍 Finding What You Need
By Role
- System Administrators: Start with Installation and Configuration
- End Users: Begin with Tool Reference and Quick Start
- Developers: Follow the Developer Journey starting with Development Setup
- Architects: Review Architecture Overview and System Design
By Task
- Setting Up: Installation → Configuration
- Using Tools: Tool Reference → Specific tool documentation
- Developing: Setup → Workflows → Code Style
- Understanding Architecture: Overview → Components → Data Flow
- Troubleshooting: Troubleshooting Guide or relevant tool documentation
By Problem Type
- Installation Issues: Installation Guide and Troubleshooting
- Configuration Problems: Configuration Guide
- Tool Behavior Questions: Specific Tool Documentation
- Development Questions: Contributing Guides
- Architecture Questions: Architecture Documentation
📝 Contributing to Documentation
This documentation follows the standards defined in CLAUDE.md:
- Accuracy: Documentation must reflect actual code behavior
- Completeness: Cover all user-facing functionality
- Accessibility: Understandable by intended audience
- Currency: Updated with every related code change
To contribute:
- Follow the Development Workflows
- Maintain Code Style Standards
- Include comprehensive Testing
- Update relevant documentation sections
Need Help? Check the Troubleshooting Guide or explore the specific documentation section for your use case. For development questions, start with the Contributing Guidelines.