tao-train-oneformer

Pass

Audited by Gen Agent Trust Hub on Sep 17, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill references a pretrained Swin Transformer backbone stored in a public GitHub repository. This is a common practice in computer vision for initializing model training with established weights.
  • [COMMAND_EXECUTION]: The skill facilitates the execution of TAO Toolkit commands (such as oneformer train, oneformer export, and oneformer inference) via a shell inside an NVIDIA Docker container. These operations are restricted to the containerized environment and are consistent with the skill's stated purpose.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 17, 2026, 02:56 AM
Security Audit — agent-trust-hub — tao-train-oneformer