A FastAPI web app for code review using Ollama when available, with a built-in local fallback analyzer.
- Paste code and choose Python, JavaScript, or C++
- Review for syntax errors, type errors, logical issues, and security issues
- Receive suggested fixes in a cleaner browser UI
- Falls back to a local analyzer when Ollama is unavailable or returns unusable output
- Uses Python AST checks,
ruff,mypy, and runtime traceback capture for stronger Python diagnostics - Shows an execution trace panel when runtime failures occur
-
Ensure Ollama is installed and the model is available:
ollama pull tinyllama
-
Install dependencies:
pip install -r requirements.txt
-
Start the app:
python3 app.py
-
Open
http://localhost:3033
The repo includes render.yaml for Render.
- Push the repository to GitHub.
- Create a new Render Blueprint from the repo.
- Render will install dependencies from
requirements.txtand startuvicorn app:app --host 0.0.0.0 --port $PORT.