tao-train-pointpillars

Pass

Audited by Gen Agent Trust Hub on Sep 17, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill consists primarily of configuration schemas and YAML templates used by the NVIDIA TAO Toolkit. It does not include standalone executable scripts or arbitrary binary files.
  • [SAFE]: External dependencies are restricted to official NVIDIA container registries (nvcr.io). These references are legitimate and originate from the trusted vendor authoring the skill.
  • [SAFE]: No patterns of prompt injection, data exfiltration, or credential harvesting were detected. Encryption keys mentioned in templates (e.g., 'tlt_encode') are standard placeholders used by the toolkit for artifact protection.
  • [SAFE]: The skill follows least-privilege principles, defining model-specific actions that operate within isolated container environments.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 17, 2026, 02:56 AM
Security Audit — agent-trust-hub — tao-train-pointpillars