tao-train-mask-auto-label

Pass

Audited by Gen Agent Trust Hub on Jun 18, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: No malicious patterns detected in instructions or scripts.
  • [CREDENTIALS_UNSAFE]: The encryption_key parameter in the schemas defaults to an empty string. While this is not a leak of a private secret, it represents a default configuration that users should be aware of when handling encrypted models as noted in the skill's own benchmark report.
  • [DATA_EXFILTRATION]: Weights & Biases (WandB) integration is enabled by default (enable: true) in the action schemas. This is a standard machine learning experiment-tracking feature but results in training metadata being sent to the WandB platform unless disabled by the user.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 18, 2026, 04:35 PM
Security Audit — agent-trust-hub — tao-train-mask-auto-label