A starter Python project for vulnerability prioritization / cybersecurity risk scoring built around the workflow described in the uploaded markdown.

1. Download and cache official vulnerability data.
2. Merge CISA KEV and NVD records into a master table.
3. Build structured and text features from CVE metadata and descriptions.
4. Train baseline and stronger models for high-priority vulnerability prediction.
5. Export interpretable results, plots, and case-study tables.

python -m src.download_data
python -m src.preprocess
python -m src.feature_engineering
python -m src.train_baseline
python -m src.train_xgboost
python -m src.anomaly_detection
python -m src.interpret

• This repository is intentionally a credible starter skeleton, not a fully polished production system.
• You will likely need to edit column names and tweak preprocessing after inspecting the real raw files.
• The code favors clarity and modularity so that each file can be debugged independently.

vulnerability-prioritization-project/
├── data/
│   ├── raw/
│   └── processed/
├── notebooks/
│   ├── 01_eda.ipynb
│   └── 02_modeling.ipynb
├── src/
│   ├── download_data.py
│   ├── preprocess.py
│   ├── feature_engineering.py
│   ├── train_baseline.py
│   ├── train_xgboost.py
│   ├── anomaly_detection.py
│   ├── interpret.py
│   └── utils.py
├── results/
│   ├── figures/
│   └── tables/
├── README.md
└── requirements.txt