tao-train-sparse4d
Pass
Audited by Gen Agent Trust Hub on Aug 25, 2026
Risk Level: SAFENO_CODECOMMAND_EXECUTION
Full Analysis
- [SAFE]: The skill is entirely informational and providing-configuration in nature. It contains no prompt injections, exfiltration patterns, or obfuscated content. All referenced resources and documentation are consistent with the primary purpose of ML model training using NVIDIA's official toolkit.
- [NO_CODE]: No executable scripts (Python, Bash, or otherwise) are bundled within the skill package. The skill folder consists of Markdown documentation, JSON schemas, and YAML templates.
- [COMMAND_EXECUTION]: While the skill documents how to execute shell commands (e.g.,
sparse4d train,annotations convert), these are standard CLI tools from the NVIDIA TAO Toolkit. The instructions provide guidance on parameters and data paths without any unauthorized privilege escalation or persistence mechanisms.
Audit Metadata