8phase

AI-native, LLM-driven threat-modeling skill (fr33d3m0n/threat-modeling v3.1.0) for automated software risk analysis, security audit, and penetration testing — covers the full OWASP MCP Top 10 (2025), the SAO (Subject-Action-Object) agent threat model, 13 pre-built MITRE ATT&CK agent attack chains, and the `SKILL.MD = UNTRUSTED` trust-inversion paradigm for agent security.

Code-First Deep Threat Modeling - LLM-native security analysis framework with automated 8-phase workflow, dual-track knowledge architecture (Security Controls + Threat Patterns), and comprehensive verification capabilities. Transform any codebase into structured threat models without design documents.

Official Animalz Claude Code plugins for content marketing workflows