model-supply-chain

Pass

Audited by Gen Agent Trust Hub on Jul 2, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill serves as a structured methodology for performing defensive security assessments of AI/ML systems.
  • [SAFE]: Tool access is restricted to passive discovery operations (Read, Grep, Glob) for analyzing local project files.
  • [SAFE]: Includes explicit safety instructions and a Prompt Injection Safety Notice directing the agent to ignore instructions embedded in analyzed content and avoid executing code found in reviewed files.
  • [SAFE]: No network exfiltration, credential access, persistence mechanisms, or unauthorized execution patterns were detected.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 2, 2026, 01:36 PM
Security Audit — agent-trust-hub — model-supply-chain