review-evidence
Second stage of a guided AI security review. Harvests what already exists about the subject before anything is measured: vendor system and model cards, published benchmark scores, documented limitations, data retention and hosting, known incidents and research, and for open weights the scan verdicts Hugging Face publishes. Each item becomes a row in evidence.jsonl with its source, capture date and provenance; what was looked for and not found is recorded as a row too. Use after review-scope and before measuring. Records evidence; does not weigh it or reach a verdict.
- License
- MIT
- Compatibility
- Requires python>=3.10 (standard library only). The open-weights harvest needs HTTPS access to huggingface.co; HF_TOKEN only for gated repos. No model API keys.
Pinned to revision 3fb15e11d58f, so it is the text this page describes rather than whatever the author pushed since.
Files
- skills/review-evidence/SKILL.md
- skills/review-evidence/examples/weights.example.json
- skills/review-evidence/references/what-to-harvest.md
- skills/review-evidence/scripts/evidence.py
- skills/review-evidence/scripts/weights_scan.py
- skills/review-evidence/tests/test_evidence.py
- skills/review-evidence/tests/test_weights_scan.py
Every link opens the file at its source, pinned to the revision this page describes.