pattern-recognition

Pass

Audited by Gen Agent Trust Hub on Jun 19, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill provides high-quality instructional content and illustrative command templates for codebase auditing without any malicious code, hidden instructions, or suspicious behaviors.
  • [SAFE]: The provided grep-based patterns are standard diagnostic tools intended for local codebase scanning and include instructions for refining results to minimize false positives.
  • [SAFE]: The skill contains explicit safety guidance regarding 'Permission tripwires', instructing the agent to respect security constraints and not attempt to bypass access denials when encountering protected resources like credential files or environment variables.
  • [SAFE]: No external dependencies, remote script downloads, or unauthorized network operations were identified across the skill's instruction set or supporting evaluation files.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 19, 2026, 04:35 AM
Security Audit — agent-trust-hub — pattern-recognition