tao-train-rtdetr
Pass
Audited by Gen Agent Trust Hub on Oct 29, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill references official NVIDIA TAO Toolkit Docker images from the NVIDIA Container Registry (nvcr.io). These are trusted resources associated with the skill author.
- [COMMAND_EXECUTION]: The skill provides example commands for running model training, distillation, and export actions within containerized environments. These commands are standard for the intended use case of the TAO Toolkit.
- [INDIRECT_PROMPT_INJECTION]: The skill defines a surface for ingesting external data (images and annotations) for model training and evaluation.
- Ingestion points: Data sources are defined for training, distillation, and evaluation actions in SKILL.md and references/skill_info.yaml.
- Boundary markers: Not explicitly present in the provided templates for dataset paths.
- Capability inventory: The skill utilizes the Bash tool to execute CLI commands within Docker containers.
- Sanitization: Input sanitization is expected to be handled by the TAO Toolkit CLI and the underlying platform.
Audit Metadata