ctf-rev

SUSPICIOUS. The skill is internally coherent and does not show credential theft, exfiltration, or installer abuse, but it is an offensive security/reverse-engineering skill for AI agents. That capability is high risk by category even without malicious data flows.