breach

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 0.90). The skill content contains extensive, actionable adversary playbooks (phishing/SMS/QR OAuth-consent evasion, SPF/DKIM/DMARC evasion techniques, supply‑chain typosquatting/postinstall attack vectors, exfiltration methods like DNS tunneling/steganography, RCE and agentic compromise patterns, prompt‑injection/jailbreak payload patterns) which are dual‑use but can be directly abused to perform credential theft, data exfiltration, backdoors, and supply‑chain compromise despite being framed for authorized red‑team use.