MySQL Agentic Copilot

An advanced MySQL Copilot that evolves from simple NL → SQL into a schema-aware, self-correcting, agentic system with performance insights and optional VS Code integration.

This repository is organized into phases:

• Phase 1 (MVP) – Backend with NL → SQL, schema introspection, and safe query execution (implemented in backend/).
• Phase 2 (Agentic) – Tool-based agent (schema, run, explain, self-correct).
• Phase 3 (Differentiator) – Query performance optimization engine and/or SQL dialect migration + VS Code integration.

Structure

mysql-agentic-copilot/
│── backend/          # FastAPI backend (Phase 1 complete)
│── README.md         # Project-level documentation

Running the Backend (Phase 1)

See backend/README.md for detailed steps.

Quick start:

cd backend
python -m venv venv
source venv/bin/activate  # Windows: venv\\Scripts\\activate
pip install -r requirements.txt
cp .env.example .env
# Add your OpenAI key in .env
uvicorn main:app --reload

API docs: http://127.0.0.1:8000/docs

Docker Usage

A Dockerfile is provided in backend/ and docker-compose.yml in the root.

To build and run with Docker:

docker compose up --build

This will spin up:

• backend service on port 8000
• (You can later add a mysql service for local DB if needed.)

Roadmap

• Phase 1 – MVP backend
• Phase 2 – Agentic execution & self-correction
• Phase 3 – Performance analysis & VS Code integration