physical-ai-datasets
Pass
Audited by Gen Agent Trust Hub on Jun 21, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill functions as a documentation and routing catalog for NVIDIA-authored datasets hosted on Hugging Face.
- [EXTERNAL_DOWNLOADS]: The skill provides instructions to download datasets from Hugging Face (
huggingface.co) and specialized tooling from official NVIDIA GitHub repositories (e.g.,github.com/NVlabs,github.com/nv-tlabs). These are well-known services and vendor-controlled resources necessary for the skill's functionality. - [COMMAND_EXECUTION]: Provides standard command-line instructions for environment setup, including package installation via
apt,pip, anduv, as well as dataset management usinggit-lfsand thehuggingface-cli(hf). These are legitimate actions for a data management skill. - [SAFE]: Includes a dedicated security section ('Verifying secrets safely') that instructs users and agents on how to check for the presence of the
HF_TOKENwithout accidentally echoing the secret value to terminal logs, demonstrating good security practice.
Audit Metadata