CodeSensei

CodeSensei gives students educational feedback on public GitHub repositories or uploaded ZIP projects. It combines deterministic security checks with an optional AI review, streams live progress, and produces a navigable report with a score, file tree, architecture notes, and actionable findings.

Start the full app

Open two PowerShell terminals.

Frontend:

npm install
npm run dev

Backend:

cd backend
py -m venv .venv
.\.venv\Scripts\Activate.ps1
pip install -r requirements.txt
uvicorn app.main:app --reload --port 8000

Then open http://localhost:5173. The frontend uses VITE_API_URL and VITE_WS_URL from .env; both default to http://localhost:8000.

Configuration

• The backend uses SQLite by default, so local development needs no database setup.
• Set DATABASE_URL to a PostgreSQL SQLAlchemy URL in production.
• Set OPENAI_API_KEY to enable the optional AI review. Static security and quality checks work without it.
• Set OPENAI_MODEL to choose the review model.
• Set FRONTEND_URL when the frontend is deployed somewhere other than http://localhost:5173.

See backend/README.md for backend details. FastAPI documentation is served at http://localhost:8000/docs.

If the backend is unavailable, the frontend deliberately enters an interactive demo report so the complete interface remains testable.

Verification

npm run lint
npm run build
cd backend
python -m unittest discover -s tests -v