FastAPI CRUD MCP

A minimal CRUD API for “items,” built with FastAPI and exposed as MCP tools via FastAPI-MCP. Includes a scenario-driven client harness using PydanticAI and Rich.

🚀 Features

• FastAPI: high-performance HTTP API

• SQLAlchemy + Pydantic: ORM models + input/output schemas

• FastAPI-MCP: auto-expose your endpoints as MCP tools (/mcp/tools, /mcp/events)

• Rich CLI: beautiful, colored terminal output for scenario runs

• Scenario Runner: client harness that drives and validates your API via PydanticAI agents

• SQLite backend for demo; easily swap to PostgreSQL, MySQL, etc.

Related MCP server: Pydantic AI Documentation Server

📦 Project Layout

⚙️ Installation & Setup

1. Clone & enter directory

   git clone https://github.com/yourusername/fastapi-crud-mcp.git cd fastapi-crud-mcp

2. Create & activate a virtualenv

   uv venv source .venv/bin/activate

3. Install dependencies

   uv sync

4. Environment variables Copy the example and adjust if needed:

   cp .env.example .env
   MCP_HOST_URL='http://127.0.0.1:8000/mcp' LLM_PROVIDER='openai' LLM_MODEL_NAME='gpt-4o-mini' LLM_MODEL=${LLM_PROVIDER}:${LLM_MODEL_NAME} OPENAI_API_KEY=sk-proj-your-api-key-here

🏃 Running the Server

• API docs → http://localhost:8000/docs

• OpenAPI JSON → http://localhost:8000/openapi.json

🤖 Running the Scenario Client

This harness will:

1. Load your .env settings

2. Spin up a PydanticAI agent against MCP_HOST_URL

3. Execute each scenario (create/list/get/update/delete)

4. Display rich panels for prompts & outputs

🚨 Notes & Tips

• Switch DB: edit backend/server/db.py for PostgreSQL or MySQL.

• Add auth: protect /mcp or /api via FastAPI dependencies.

• Extend scenarios: drop new entries into backend/client/scenarios.py.

• Production: add Alembic for migrations, and monitor with Prometheus.

🤝 Contributing

1. Fork 🔱

2. Create a feature branch:

   git checkout -b feature/my-feature

3. Commit & push:

   git commit -am "Add awesome feature" git push origin feature/my-feature

4. Open a PR and we’ll review!

📄 License

This project is MIT-licensed—see the LICENSE file for details.