💀 反AI蒸馏防护工具箱 | Anti-Distillation Defense Toolkit — 人可以被优化，但灵魂不能被窃取 | People can be optimized, but souls cannot be stolen

Real-time trustworthiness evaluation and safety interception for AI agents. Semantic analysis, safe alternative suggestions, multi-step attack chain detection, and LLM-as-Judge.

Skills Catalog for Codex

The safety net for vibe coding — auto-snapshot, architecture guard, and smart recovery for AI-assisted development.