Structured error-to-rule learning system for AI agents. Activate when an agent makes a mistake, receives a correction from the user, or needs to check past lessons before making a decision. Converts errors into executable rules (not reflections) stored in lessons.md, and enforces pre-decision rule scanning to prevent repeat mistakes. Supports sharing anonymized lessons to a community repository.

面向AI代理的结构化错误转规则学习系统。当代理出错、收到用户纠正或决策前需要回顾过往课程时启动。将错误转化为可执行规则（非反思性内容）并存储至lessons.md文件中，通过决策前的规则扫描机制防止重复错误。支持将匿名化课程分享至社区仓库。