langgraph-fastapi-mcp-server-demo
このプロジェクトは、FastAPIを使用してショッピングリストアプリをMCPサーバーに変換し、LangGraphベースのチャットボットと接続するデモです。ユーザーはチャットを通じてショッピングリストを管理できます。MCPプロトコルを利用することで、AIモデルと外部データソースとのシームレスなインタラクションが可能です。
GitHubスター
0
ユーザー評価
未評価
フォーク
0
イシュー
0
閲覧数
0
お気に入り
0
README
LangGraph FastAPI MCP Server Demo
This is a sample project that turns a shopping list FastAPI app into an MCP server and connects a LangGraph based chatbot to the MCP server so that it can manage the user's shopping list via chat.
Technology Stack
- FastAPI: a modern, fast (high-performance), web framework for building APIs with Python 3
- FastAPI-MCP: an open-source library that exposes your FastAPI endpoints as Model Context Protocol (MCP) tools (with Auth)
- MCP (Model Context Protocol): an open protocol that facilitates seamless interaction between AI models and external data sources or tools
- LangChain: An opens-source framework that helps developers build applications powered by large language models (LLMs)
- LangGraph: An opens-source framework by LangChain for building and managing complex AI agents using graph-based architecture
- ChatGPT: A popular LLM provided by OpenAI
- Gradio: A popular opens-source library for building web interfaces for machine learning models. This project uses Gradio to create a web interface for the chatbot.
- LangSmith: An opens-source framework by LangChain for tracing and monitoring LLM applications.
- uv: A fast opens-source Python package manager - a drop-in replacement for
pip,condaandvirtualenv.
Quick Start
1. Install uv
Follow the instructions here to install uv, if you haven't already.
2. Install Dependencies
uv sync
3. Set Up OpenAI API Key
Create a .env file in the project root:
# .env
OPENAI_API_KEY=your-openai-api-key-here
4. Run the FastAPI app
uv run uvicorn server.main:app --host 0.0.0.0 --port 8000 --reload
Try the FastAPI app at http://localhost:8000/docs
5. Run the Chatbot
uv run chatbot.py
The chatbot will be available at http://localhost:7860
Usage Examples
Try these example queries in the chatbot:
- "I need to buy a watermelon",
- "What's in my shopping list?",
- "Let's add spaghetti and tomato sauce",
- "I just bought the watermelon",
- "Remove the tomato sauce",
- "I need an extra spaghetti",
Dependencies
- LangGraph: For creating the React agent
- Gradio: For the chat interface
- LangChain OpenAI: For the language model integration
- LangChain Core: For tool definitions
- FastAPI: For the MCP server
- SQLAlchemy: For database operations
License
This project is open source and available under the MIT License.