tao-train-action-recognition
Pass
Audited by Gen Agent Trust Hub on Sep 17, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill implements standard NVIDIA TAO (Train, Adapt, Optimize) workflows for video action classification. All operations, including training, evaluation, and inference, are designed to run within Docker containers, which provides an isolated execution environment.
- [SAFE]: Remote resources, specifically the Docker image 'nvcr.io/nvstaging/tao/tao-toolkit-pyt:7.2.0-rc-36-multiarch', are hosted on the official NVIDIA Container Registry (NVCR), a trusted vendor-controlled infrastructure.
- [SAFE]: The skill follows standard practices for data handling by mounting local directories (data, specs, results) to the container. No sensitive host paths (e.g., SSH keys, cloud credentials) are accessed or exposed.
- [SAFE]: The presence of a digital signature ('skill.oms.sig') in a standard Sigstore/In-toto format indicates a commitment to skill integrity and authenticity.
Audit Metadata