Community生产力与协作github.com

yanhongxi-openclaw/proactive-self-improving-agent

自动捕获经验并安全进化的技能。触发条件:(1)命令/操作失败时→记ERRORS.md (2)被用户纠正('不对'/'应该是')时→记LEARNINGS.md (3)用户需要不存在的能力时→记FEATURE_REQUESTS.md (4)外部API/工具出错时→记ERRORS.md (5)发现自己知识过时/错误时→记LEARNINGS.md (6)发现更好做法时→记LEARNINGS.md (7)每个任务完成时→回顾过程,有新经验则记LEARNINGS.md。去重原则:如果没有新经验或已有条目已覆盖则跳过不写。每次写入同时在.learnings/CHANGELOG.md追加JSONL日志。经验反复出现≥3次时晋升到AGENTS.md/TOOLS.md/SOUL.md。详见正文。

proactive-self-improving-agent 是什么?

proactive-self-improving-agent is a Claude Code agent skill that 自动捕获经验并安全进化的技能。触发条件:(1)命令/操作失败时→记ERRORS.md (2)被用户纠正('不对'/'应该是')时→记LEARNINGS.md (3)用户需要不存在的能力时→记FEATURE_REQUESTS.md (4)外部API/工具出错时→记ERRORS.md (5)发现自己知识过时/错误时→记LEARNINGS.md (6)发现更好做法时→记LEARNINGS.md (7)每个任务完成时→回顾过程,有新经验则记LEARNINGS.md。去重原则:如果没有新经验或已有条目已覆盖则跳过不写。每次写入同时在.learnings/CHANGELOG.md追加JSONL日志。经验反复出现≥3次时晋升到AGENTS.md/TOOLS.md/SOUL.md。详见正文。.

兼容平台~Claude Code~Codex CLI~Cursor
npx skills add https://github.com/yanhongxi-openclaw/proactive-self-improving-agent/tree/main/skills/proactive-self-improving-agent

Installed? Explore more 生产力与协作 skills: steipete/gemini, steipete/gh-issues, steipete/skill-creator · View all 6 →

在你喜欢的 AI 中提问

打开一个已预加载此 Agent Skill 的新对话。

文档

proactive-self-improving-agent 是做什么的?

Automatically captures experiences and safely evolves. Trigger conditions: (1) command/operation failure → record in ERRORS.md (2) user correction ('no'/'should be') → record in LEARNINGS.md (3) user requests missing capability → record in FEATURE_REQUESTS.md (4) external API/tool error → record in ERRORS.md (5) own knowledge outdated/incorrect → record in LEARNINGS.md (6) better practice discovered → record in LEARNINGS.md (7) each task completion → review process, if new experience exists record in LEARNINGS.md. Dedup rule: skip if no new experience or existing entry already covers it. Each write also appends a JSONL log to .learnings/CHANGELOG.md. Experiences recurring ≥3 times promote to AGENTS.md/TOOLS.md/SOUL.md. See body for details.

相关技能