tao-train-mask2former
Pass
Audited by Gen Agent Trust Hub on Jun 18, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADS
Full Analysis
- [SAFE]: The skill documentation and configuration files (schemas and templates) follow standard machine learning workflow patterns for the TAO Toolkit. Static analysis found no evidence of prompt injection attempts, unauthorized access to sensitive files, or data exfiltration mechanisms.
- [EXTERNAL_DOWNLOADS]: The skill references external downloads for pre-trained Swin Transformer weights from a specific GitHub repository and provides instructions for using Docker images from the official NVIDIA Container Registry (nvcr.io). These actions are appropriate for the skill's function and originate from well-known domains.
Audit Metadata