🚀 MCP FastAPI Docs

Intelligent FastAPI Documentation Search via Model Context Protocol (MCP)

A standalone MCP server that provides semantic search, topic exploration, and code ex## � Documentation

📚 Complete Documentation - Full documentation with table of contents

Quick Links

• Getting Started

  • Installation - Set up your environment
  • Initial Setup - Populate the cache
  • Quick Start - Run your first queries

• Integration Guides

  • LLM Quick Start - Integrate with AI assistants
  • Claude Desktop - Claude setup guide
  • VS Code Setup - VS Code configuration

• Development

  • Testing Guide - Run and write tests
  • Architecture - Understand the codebase
  • Contributing - How to contributextraction for FastAPI documentation. Perfect for AI assistants like Claude, Copilot, and other LLM-powered tools.

---

🌟 Features

• � Semantic Search - Natural language queries powered by sentence transformers
• 📚 Topic Listing - Browse all 104 FastAPI documentation topics
• 📖 Topic Retrieval - Get full content for any specific topic
• 💻 Code Examples - Extract code snippets with language and syntax highlighting
• 🔄 Version Comparison - Compare features across different FastAPI versions
• ⚡ Fast & Local - Runs entirely on your machine with ChromaDB vector storage
• 🎯 Zero Config - Works out of the box with sensible defaults

---

🎯 Quick Start

Prerequisites

• Python 3.10 or higher
• uv (recommended) or pip

Installation

# Clone the repository
git clone https://github.com/yourusername/mcp-fastapi-docs.git
cd mcp-fastapi-docs

# Install dependencies with uv (recommended)
uv sync

# Or with pip
pip install -e .

Initial Setup

Populate the cache with FastAPI documentation:

uv run python scripts/populate_cache.py

This downloads FastAPI docs and builds the vector database (takes 2-3 minutes).

---

� Usage

1. As an MCP Server (Recommended)

Configure your MCP client (Claude Desktop, VS Code, etc.):

{
  "mcpServers": {
    "fastapi-docs": {
      "command": "uv",
      "args": ["run", "mcp-fastapi-docs"],
      "env": {
        "CACHE_DIR": "/absolute/path/to/cache",
        "LOG_LEVEL": "INFO"
      }
    }
  }
}

See LLM_QUICK_START.md for detailed integration guides.

2. Direct Python API

import asyncio
from mcp_fastapi_docs.server.tools import search_docs, list_topics, get_topic

async def main():
    # Search documentation
    results = await search_docs("how to use path parameters", version="latest")

    # List all topics
    topics = await list_topics(version="latest")

    # Get specific topic
    content = await get_topic("Path Parameters", version="latest")

asyncio.run(main())

3. Interactive Testing

# Manual testing
uv run python scripts/test_mcp_manual.py

# Interactive CLI
uv run python scripts/test_interactive.py

---

🛠️ Available Tools

The MCP server exposes 6 powerful tools:

Tool	Description	Parameters
search_docs	Semantic search across documentation	query, version, max_results
list_versions	List all available documentation versions	None
list_topics	List all available topics	version
get_topic	Get full content for a topic	topic, version
get_example	Extract code examples for a feature	feature, version
compare_versions	Compare a feature across versions	feature, versions

All tools support version="latest" which automatically resolves to the newest available version.

---

📁 Project Structure

mcp-fastapi-docs/
├── src/mcp_fastapi_docs/
│   ├── core/           # Cache, fetcher, processor
│   ├── models/         # Pydantic models & config
│   ├── search/         # Semantic search engine
│   ├── server/         # MCP server & tool registry
│   └── utils/          # Logging utilities
├── scripts/            # Setup & testing scripts
├── cache/              # Local documentation & embeddings
├── docs/               # Documentation guides
├── tests/              # Pytest test suite
└── examples/           # Usage examples

---

🧪 Testing

# Run all tests
uv run pytest

# Run with coverage
uv run pytest --cov=src/mcp_fastapi_docs --cov-report=html

# Test specific functionality
uv run python scripts/test_search.py

See TESTING.md for comprehensive testing guide.

---

🔌 VS Code Integration

This project includes VS Code MCP server configuration:

1. Open workspace in VS Code
2. MCP server auto-registers via .vscode/mcp.json
3. Use Copilot or other MCP extensions to query FastAPI docs

See .vscode/README_MCP.md for details.

---

🎓 Learning & Extending

This project is designed as a learning template for building MCP servers:

• Self-contained - All components in one repo
• Well-documented - Extensive inline comments & guides
• Modular design - Easy to adapt for other frameworks
• Best practices - Type hints, async/await, proper error handling

Want to build similar servers for Flask or Django? This codebase serves as a complete reference implementation.

---

� Documentation

• LLM Quick Start - Get started with LLM integration
• LLM Integration Guide - Detailed integration patterns
• Quick Test Guide - Testing walkthrough
• Testing Guide - Comprehensive test documentation
• VS Code Setup - VS Code configuration

---

🤝 Contributing

Contributions are welcome! This project follows standard open-source practices:

1. Fork the repository
2. Create a feature branch (git checkout -b feature/amazing-feature)
3. Commit your changes (git commit -m 'Add amazing feature')
4. Push to the branch (git push origin feature/amazing-feature)
5. Open a Pull Request

---

🐛 Troubleshooting

Issue: No FastAPI documentation versions available

• Solution: Run uv run python scripts/populate_cache.py

Issue: Import errors or missing dependencies

• Solution: Run uv sync or pip install -e .

Issue: MCP server not responding

• Solution: Check logs in cache/logs/mcp-fastapi-docs.log

---

📄 License

MIT License - see LICENSE for details.

---

🙏 Acknowledgments

• FastAPI - The amazing framework this serves
• Model Context Protocol - The protocol powering AI integration
• ChromaDB - Vector database for semantic search
• Sentence Transformers - Embedding models

---

🚀 Future Plans

• Support for multiple FastAPI versions
• Incremental cache updates
• Full-text search alongside semantic search
• Web UI for documentation browsing
• API rate limiting and caching strategies

---

Built with ❤️ for the FastAPI and AI community