tao-train-foundation-stereo

Pass

Audited by Gen Agent Trust Hub on Jun 23, 2026

Risk Level: SAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS]: Fetches pre-trained model weights from Hugging Face for model initialization in the training workflow.
  • [COMMAND_EXECUTION]: Instructs the agent to execute the depth_net CLI tool within Docker containers for standard machine learning tasks like training, quantization, and evaluation. This behavior is consistent with the skill's primary purpose.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied annotation files and image data which represent a potential ingestion surface.
  • Ingestion points: The skill reads annotation files (annotations.txt) and image directories in SKILL.md and through various specification templates.
  • Boundary markers: Absent.
  • Capability inventory: Includes the ability to execute shell commands via the Bash tool to run TAO Toolkit actions.
  • Sanitization: Absent.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 23, 2026, 12:10 PM
Security Audit — agent-trust-hub — tao-train-foundation-stereo