tao-train-rtdetr
Pass
Audited by Gen Agent Trust Hub on Sep 17, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill provides standard architectural guidelines and configuration schemas for RT-DETR model workflows. All operations described involve the use of the NVIDIA TAO Toolkit in a containerized environment.
- [SAFE]: External resource references, such as container images hosted on nvcr.io (NVIDIA Container Registry), are from trusted vendor-owned domains and are appropriate for the skill's stated purpose.
- [SAFE]: Instructions for data handling (S3 paths, dataset image directories) and model checkpoints follow standard AI development practices and do not exhibit data exfiltration or credential exposure patterns.
- [SAFE]: The signature file (skill.oms.sig) and benchmarking documentation (BENCHMARK.md) confirm the skill's origin and evaluation history, aligning with secure distribution practices.
Audit Metadata