triage-validation

SUSPICIOUS: the skill is internally coherent and does not show malicious install, credential theft, or third-party data routing, but it equips an AI agent with offensive security validation workflows for real targets. That makes it materially risky even though it is not malware.