Security audit framework for AI agent skills, MCP servers, and packages. Your LLM does the analysis — we provide structure, prompts, and a shared trust database.
Major overhaul: LLM-based audits, peer review, provenance chains, anti-collusion, AI-specific detection patterns