tao-train-single-step
Pass
Audited by Gen Agent Trust Hub on Sep 17, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill dictates a standard instructional workflow for model training. The metadata correctly maps to the vendor (NVIDIA Corporation). It uses local evaluation scripts (
scripts/list_tao_models.py,scripts/resolve_tao_image.py) which operate within the agent framework context without external network hazards or untrusted data execution paths.
Audit Metadata