tao-train-single-step
Pass
Audited by Gen Agent Trust Hub on Sep 22, 2026
Risk Level: SAFE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests data from external sources such as training and evaluation dataset URIs, which provides a surface for indirect prompt injection. Ingestion points:
train_dataset_uri,eval_dataset_uri. Capability inventory: Bash and Write tools. Boundary markers: Absent. Sanitization: Not described in instructions. - [EXTERNAL_DOWNLOADS]: The skill references the official NVIDIA TAO Skill Bank on GitHub and official documentation.
- [COMMAND_EXECUTION]: The workflow involves executing local scripts such as
list_tao_platforms.pyandresolve_tao_image.pyto configure the training environment.
Audit Metadata