redteam-mindset

SUSPICIOUS/HIGH-RISK vulnerable skill, not confirmed malware. Its purpose and content are internally aligned as an offensive red-team methodology guide, but it materially increases an AI agent's capability to conduct autonomous penetration testing, use OOB infrastructure, install extra tooling, and chain into more offensive skills. No clear credential-harvesting or covert exfiltration is embedded, yet the operational footprint is inherently high risk for AI-agent use.