qe-pentest-validation

SUSPICIOUS/HIGH-RISK skill. The stated purpose matches the capabilities, so this is not deceptive malware, but it is an offensive-security skill that enables AI-driven exploitation against live systems. Safeguards are described, yet the footprint includes high-impact actions like reverse-shell proof, cookie theft PoC, metadata extraction, and credential-stuffing validation, plus delegation to other skills. No clear malicious exfiltration path is shown, but the capability itself is dangerous and disproportionate for general agent use.