check-invariants

Pass

Audited by Gen Agent Trust Hub on Jul 26, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill's primary function is architectural review, involving reading local documentation (e.g., docs/architecture/INDEX.md) and project files (design.md, git diff) to evaluate compliance with internal rules.
  • [SAFE]: No network operations, data exfiltration patterns, or unauthorized sensitive file access were detected.
  • [SAFE]: The skill does not utilize external dependencies, remote code execution, or any forms of obfuscation.
  • [SAFE]: There are no signs of prompt injection or attempts to bypass agent safety guidelines; the instructions focus on structured reasoning and reporting.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 26, 2026, 10:10 AM
Security Audit — agent-trust-hub — check-invariants