security-audit

SUSPICIOUS: the skill is internally consistent with its stated purpose, but that purpose is to give an AI agent security-audit/offensive-review capability and to delegate to another unseen agent. No obvious credential harvesting or exfiltration is present, yet the combination of autonomous security scanning, untrusted code ingestion, shell access, and transitive agent trust makes the overall risk high enough to treat as a high-risk security tool rather than benign documentation.