GEMINI.md

Repo Guidelines

This repository contains the LLM-based Cancer Risk Assessment Assistant.

Core Technologies

• FastAPI for the web framework
• LangChain for LLM orchestration
• uv for environment and dependency management
• hydra: for configuration management

Coding Philosophy

• Prioritize clarity and reusability.
• Favor simple replication over heavy abstraction.
• Keep comments short and only where the code isn't self-explanatory.
• Avoid verbose docstrings for simple functions.

Testing

• Write meaningful tests that verify core functionality and prevent regressions.
• Run tests with uv run pytest.

Development Setup

• Create the virtual environment (at '.venv') with uv sync.

Running commands

• As the repository uses uv, the uv should be used to run all commands, e.g., "uv run python ..." NOT "python ...".

These guidelines apply to the entire repository. A multi-page Streamlit interface for expert feedback can be launched with uv run streamlit run apps/streamlit_ui/main.py. The first page, User Profile, allows experts to load or create a profile stored in st.session_state.user_profile. The second page, Configuration, lets experts choose the model and knowledge base modules while previewing the generated prompt. The third page, Assessment, runs the AI analysis, displays a results dashboard, and provides export and chat options.

Important Note for Developers

When making changes to the project, ensure that the following files are updated to reflect the changes:

• README.md
• AGENTS.md
• GEMINI.md

Risk Model Availability

Risk calculators exposed to Gemini-based agents include:

• Gail - Breast cancer risk
• Claus - Breast cancer risk based on family history
• Tyrer-Cuzick - Breast cancer risk (IBIS model)
• BOADICEA - Breast and ovarian cancer risk (via CanRisk API)
• PLCOm2012 - Lung cancer risk
• LLPi - Liverpool Lung Project improved model for lung cancer risk (8.7-year prediction)
• CRC-PRO - Colorectal cancer risk
• PCPT - Prostate cancer risk
• Extended PBCG - Prostate cancer risk (extended model)
• Prostate Mortality - Prostate cancer-specific mortality prediction
• MRAT - Melanoma risk (5-year prediction)
• aMAP - Hepatocellular carcinoma (liver cancer) risk
• QCancer - Multi-site cancer differential

Register additional models in src/sentinel/risk_models/__init__.py so they are available system-wide.