Agent Skills: red-teaming

Comprehensive red teaming methodology for both cybersecurity and AI/LLM systems. Use when conducting adversary emulation, vulnerability assessment, attack simulation, or security validation. Trigger on requests for penetration testing, threat modeling, security audits, MITRE ATT&CK operations, LLM safety testing, prompt injection attacks, or compliance validation (OWASP, NIST, TIBER, DORA, EU AI Act). Apply when users ask to "test like an attacker", "red team our system", "validate security posture", "assess LLM vulnerabilities", or "simulate cyber attacks". Includes planning frameworks, execution strategies, reporting templates, and progressive references to specialized attack techniques and tools.