{"skill":{"slug":"skill-autogenesis","displayName":"Skill Autogenesis","summary":"Review completed work, summarize reusable procedures, identify recurring workflow patterns, and decide whether to create a skill, patch an existing skill, st...","tags":{"automation":"1.3.2","hermes":"1.3.2","latest":"1.3.2","openclaw":"1.3.2","procedural-memory":"1.3.2","skills":"1.3.2","summarization":"1.3.2"},"stats":{"comments":0,"downloads":111,"installsAllTime":0,"installsCurrent":0,"stars":0,"versions":3},"createdAt":1776269364408,"updatedAt":1776421611407},"latestVersion":{"version":"1.3.2","createdAt":1776421556345,"changelog":"# Changelog\n\n## 1.3.2\n\n- Added hard stop rules that explicitly block invalid `skill_manage(create)` usage.\n- Tightened lifecycle rules so only `knowledge_type=procedure` may enter procedural skill actions.\n- Added `references/hard-stop-rules.md` with a fast test for valid skill candidates.\n- Updated the README to surface hard interception behavior.\n\n## 1.3.1\n\n- Added a mandatory internal classification template before any persistence action.\n- Added a checklist gate that must fully pass before create or patch is allowed.\n- Extended verification rules so create and patch require a completed decision record.\n- Updated the README to emphasize classification template first, file writes second.\n\n## 1.3.0\n\n- Added an explicit decision matrix to route outcomes to skill, memory, prompt, or no-op.\n- Added an output contract so agents classify the result before writing files.\n- Added a recurrence warning that repeated rules still do not qualify as skills without executable procedure.\n- Added `references/classification-examples.md` with concrete examples of what should and should not become a skill.\n\n## 1.2.2\n\n- Changed the default posture from implied creation to explicit classification first.\n- Added a hard gate that only executable reusable procedures may become skills.\n- Added routing guidance for non-skill outcomes such as preferences, policies, boundaries, and prompt-level governance.\n- Added a pitfall warning that rules are not skills unless they contain trigger, action, and verification.\n- Updated the bilingual README to reduce the common misread that this skill should create a new rule after every success.\n\n## 1.2.1\n\n- Refined positioning to emphasize verification-gated skill creation rather than unconditional autonomous writes.\n- Clarified that skill creation happens only when recurrence, stability, and environment policy permit it.\n- Tightened README wording to better communicate lifecycle controls and safety boundaries.\n- Kept local fallback reference behavior for GitHub sources.\n\n## 1.2.0\n\n- Added explicit source resolution policy: GitHub first, local fallback second, `[UNVERIFIED]` last.\n- Added local fallback reference files under `references/fallback/`.\n- Added `skill_manage`-style lifecycle handling for create, patch, edit, write_file, remove_file, and guarded delete.\n- Added reusable template file for generated skills.\n\n## 1.1.0\n\n- Added lifecycle-oriented behavior modeled after `skill_manage`.\n- Added support for supporting-file management and duplicate-skill avoidance.\n\n## 1.0.0\n\n- Initial release.\n- Added automatic workflow distillation, recurrence detection, and skill generation guidance.","license":"MIT-0"},"metadata":{"os":null,"systems":null},"owner":{"handle":"codeblackhole1024","userId":"s177mgnzkjxtm73sg9mgwxwc9n83nygc","displayName":"codeblackhole","image":"https://avatars.githubusercontent.com/u/39405686?v=4"},"moderation":null}