tao-train-centerpose
Pass
Audited by Gen Agent Trust Hub on Jun 23, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill provides instructions and configuration templates for managing CenterPose models using official NVIDIA TAO Toolkit CLI tools and container images from the vendor's registry (nvcr.io).
- [EXTERNAL_DOWNLOADS]: The documentation references official NVIDIA container images to facilitate model deployment and execution environment setup.
- [CREDENTIALS_UNSAFE]: The JSON schemas and spec templates include an encryption_key parameter. In the provided files, this field is populated with empty strings or the standard default value (tlt_encode), which is consistent with the tool's intended use for model encryption and does not involve the exposure of private secrets.
Audit Metadata