This is a uv-managed Python project scaffold using Streamlit and LangGraph.
- Python 3.10+
 - uv installed
 
- Create the virtual environment (uv uses venvs):
 
uv venv
- Activate it:
 
- macOS/Linux (bash/zsh): 
source .venv/bin/activate - Windows (PowerShell): 
.venv\\Scripts\\Activate.ps1 
- Install dependencies:
 
uv sync
- Run Streamlit app (example):
 
streamlit run app.py
Use the utility script to file a grievance without the Streamlit UI:
uv run python scripts/add_grievance.py 3 Build pipeline flakes on Mondays
The first argument is the level (1-5) and everything after it is treated as the message. Ensure .env contains a valid OPENAI_API_KEY before running the script so the LangGraph workflow can call OpenAI.
- uv manages dependencies via 
pyproject.tomland creates a local.venv/by default when you runuv venv. - Secrets for Streamlit should go in 
.streamlit/secrets.toml(already gitignored).