tao-train-segformer
Pass
Audited by Gen Agent Trust Hub on Jun 30, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill does not exhibit any malicious behavior or safety violations. It is a legitimate wrapper for the NVIDIA TAO Toolkit.
- [COMMAND_EXECUTION]: The skill specifies commands to be executed within Docker containers, such as
segformer trainandsegformer export. These are documented vendor-provided tools for model lifecycle management. - [DATA_EXPOSURE]: While the skill templates include an
encryption_keyfield with the valuetlt_encode, this is a publicly documented default for the TAO Toolkit and is not a private credential.
Audit Metadata