[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M6F9ZAFACH7XZ8KMFK9TKARX":3,"skill-same:01M6F9ZAFACH7XZ8KMFK9TKARX":69},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":16,"categories":17,"keywords":23,"publishedAt":35,"linkOnly":36,"markdownContent":37,"markdownBody":38,"frontMatter":39,"refName":50,"commitSha":51,"skillDirectory":52,"forksCount":49,"repositoryHtmlUrl":53,"repositoryHomepageUrl":6,"ownerHtmlUrl":54,"ownerBlogUrl":6,"ownerDescription":6,"updatedAt":55,"install":56,"resources":68},"01M6F9ZAFACH7XZ8KMFK9TKARX","skill-refinement",null,"Skill 사용 후 개선 제안. ledger 분석으로 어느 skill 이 자주 실패 \u002F open_question 다발 \u002F 비효율인지 식별. Chief 가 retrospective 후 호출.","Adelie-Squad\u002Fsolosquad","Adelie-Squad","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F269218606?v=4","01M6CHETAA89FK6KREADPVEF4T","SaPN","01M6CHETE7QEKZW6FK52137ZBP",25,"MIT","https:\u002F\u002Fgithub.com\u002FAdelie-Squad\u002Fsolosquad\u002Fblob\u002Fmain\u002Fskills\u002Fskill-refinement\u002FSKILL.md",[18],{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000023","Skill与智能体开发","SCENARIO","使用场景",[24,25,26,27,28,29,30,31,32,33,34],"skill refinement","refinement","retrospective","agile","open","leader","question","chief","false","ledger","true","2026-11-10 10:12:30",false,"---\nname: skill-refinement\ndescription: Skill 사용 후 개선 제안. ledger 분석으로 어느 skill 이 자주 실패 \u002F open_question 다발 \u002F 비효율인지 식별. Chief 가 retrospective 후 호출.\nschema_version: 2\ntier: leader\nteam: _skill\ncategory: agile\nused_by: [\"chief\"]\ndev_capability: false\ntriggers:\n  keyword: [\"skill 개선\", \"skill refinement\", \"skill 평가\"]\npm_conventions:\n  anti_sycophancy: true\n  hard_gate: false\n  post_labeling: true\n  minimum_approaches: 1\n---\n\n# Skill Refinement Skill\n\n> Chief 자가학습 루프의 일부. 어떤 skill 이 잘 작동하고, 어떤 skill 이 개선 필요한지 평가.\n\n## 입력\n\n- 평가 대상 period\n- `\u003Corg>\u002Fmemory\u002Fledger\u002F\u003Ctask-id>.jsonl` — skill 호출 trace\n- `\u003Corg>\u002Fmemory\u002Fopen-questions\u002F\u003Cid>.json` — 미해결\u002Fblocking 비율\n- `\u003Corg>\u002Fmemory\u002Fagent-costs.jsonl` — skill 별 cost\n\n## 평가 차원\n\n```yaml\nper_skill_metrics:\n  skill_name: \u003Cname>\n  invocation_count: \u003CN>\n  avg_duration_ms: \u003CN>\n  avg_cost_usd: \u003CN>\n  hard_gate_failure_rate: \u003C%>      # gate 통과 못 한 비율\n  open_question_rate: \u003C%>           # 호출당 발생한 open_question 수\n  rework_rate: \u003C%>                  # 같은 task 에서 재호출된 비율\n  confidence_delta_avg: \u003C+\u002F->       # 호출 전후 confidence 변화\n```\n\n## Failure Patterns\n\n```yaml\npatterns:\n  - skill: problem-definition\n    issue: \"TDCC P4 단계 unknown 필드 too frequent (62%)\"\n    root_cause_hypothesis: \"archive context 너무 짧음 (avg N=4 entries)\"\n    fix_candidates:\n      - \"기본 archive window 확대 (4→8 entries)\"\n      - \"TDCC P4 reasoning step skip threshold 도입\"\n```\n\n## 출력\n\n```yaml\nperiod: \"...\"\ntop_3_high_friction_skills: [...]\ntop_3_efficient_skills: [...]\nrefinement_proposals:\n  - skill: \"...\"\n    proposal: \"...\"\n    expected_impact: \"...\"\n    effort: small | medium | large\n    approval_required: founder | chief_auto\n```\n\n## Anti-Sycophancy\n\n- ❌ \"skill 들이 잘 작동합니다\"\n- ✅ \"problem-definition 의 P4 단계가 62% 빈도로 unknown 필드 발생. archive context 확대로 개선 가능 (예상 -30%).\"\n\n## Reference\n\n- v1.1 PRD §5.2 RETROSPECT + §1.1 #7 (자가학습 메커니즘)\n","# Skill Refinement Skill\n\n> Chief 자가학습 루프의 일부. 어떤 skill 이 잘 작동하고, 어떤 skill 이 개선 필요한지 평가.\n\n## 입력\n\n- 평가 대상 period\n- `\u003Corg>\u002Fmemory\u002Fledger\u002F\u003Ctask-id>.jsonl` — skill 호출 trace\n- `\u003Corg>\u002Fmemory\u002Fopen-questions\u002F\u003Cid>.json` — 미해결\u002Fblocking 비율\n- `\u003Corg>\u002Fmemory\u002Fagent-costs.jsonl` — skill 별 cost\n\n## 평가 차원\n\n```yaml\nper_skill_metrics:\n  skill_name: \u003Cname>\n  invocation_count: \u003CN>\n  avg_duration_ms: \u003CN>\n  avg_cost_usd: \u003CN>\n  hard_gate_failure_rate: \u003C%>      # gate 통과 못 한 비율\n  open_question_rate: \u003C%>           # 호출당 발생한 open_question 수\n  rework_rate: \u003C%>                  # 같은 task 에서 재호출된 비율\n  confidence_delta_avg: \u003C+\u002F->       # 호출 전후 confidence 변화\n```\n\n## Failure Patterns\n\n```yaml\npatterns:\n  - skill: problem-definition\n    issue: \"TDCC P4 단계 unknown 필드 too frequent (62%)\"\n    root_cause_hypothesis: \"archive context 너무 짧음 (avg N=4 entries)\"\n    fix_candidates:\n      - \"기본 archive window 확대 (4→8 entries)\"\n      - \"TDCC P4 reasoning step skip threshold 도입\"\n```\n\n## 출력\n\n```yaml\nperiod: \"...\"\ntop_3_high_friction_skills: [...]\ntop_3_efficient_skills: [...]\nrefinement_proposals:\n  - skill: \"...\"\n    proposal: \"...\"\n    expected_impact: \"...\"\n    effort: small | medium | large\n    approval_required: founder | chief_auto\n```\n\n## Anti-Sycophancy\n\n- ❌ \"skill 들이 잘 작동합니다\"\n- ✅ \"problem-definition 의 P4 단계가 62% 빈도로 unknown 필드 발생. archive context 확대로 개선 가능 (예상 -30%).\"\n\n## Reference\n\n- v1.1 PRD §5.2 RETROSPECT + §1.1 #7 (자가학습 메커니즘)\n",{"name":5,"team":40,"tier":29,"used_by":41,"category":27,"triggers":42,"description":7,"dev_capability":36,"pm_conventions":46,"schema_version":49},"_skill",[31],{"keyword":43},[44,24,45],"skill 개선","skill 평가",{"hard_gate":36,"post_labeling":47,"anti_sycophancy":47,"minimum_approaches":48},true,1,2,"main","37b891996f0e5d7317b2abdd6e2a8acccdc94dff","skills\u002Fskill-refinement","https:\u002F\u002Fgithub.com\u002FAdelie-Squad\u002Fsolosquad","https:\u002F\u002Fgithub.com\u002FAdelie-Squad","2026-11-12 01:51:16",{"source":8,"skillName":5,"commands":57,"prompt":67,"zipAvailable":47},[58,61,64],{"tool":59,"command":60},"npx","npx skills add Adelie-Squad\u002Fsolosquad --skill skill-refinement",{"tool":62,"command":63},"bunx","bunx skills add Adelie-Squad\u002Fsolosquad --skill skill-refinement",{"tool":65,"command":66},"pnpm","pnpm dlx skills add Adelie-Squad\u002Fsolosquad --skill skill-refinement","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FAdelie-Squad\u002Fsolosquad\u002Fblob\u002Fmain\u002Fskills\u002Fskill-refinement\u002FSKILL.md",[],[70,86,100,120,141,155],{"skillNo":71,"name":72,"nameZh":6,"description":73,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":74,"categories":75,"keywords":76,"publishedAt":35},"01M6F9ZX66ETTM8NX73Z79MA21","xyz-hypothesis","검증 가능한 가설 프레임워크 — \"[X%] of [Y segment] will [Z behavior] within [T period], because [R]\". 입장+반증 조건 강제, 최소 2개 접근, V\u002FU\u002FV\u002FF 가정 분류. v1.3.5 에서 problem-definition P5 를 독립 skill 로 분리.","https:\u002F\u002Fgithub.com\u002FAdelie-Squad\u002Fsolosquad\u002Fblob\u002Fmain\u002Fskills\u002Fxyz-hypothesis\u002FSKILL.md",[],[77,78,79,80,81,34,82,83,32,31,84,29,85],"xyz hypothesis","hypothesis","product","v1","manager","segment","problem","behavior","definition",{"skillNo":87,"name":88,"nameZh":6,"description":89,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":90,"categories":91,"keywords":93,"publishedAt":35},"01M6F9ZSHR17J5CH16AQAK0ER0","workflow-refinement","Workflow 종료 시 개선 제안. stage 의존성 \u002F handoff 효율 \u002F 평균 cycle time 분석. Chief 가 RETROSPECT 에서 호출.","https:\u002F\u002Fgithub.com\u002FAdelie-Squad\u002Fsolosquad\u002Fblob\u002Fmain\u002Fskills\u002Fworkflow-refinement\u002FSKILL.md",[92],{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},[94,95,25,27,34,32,96,97,98,31,29,99],"workflow","workflow refinement","stage","time","cycle","handoff",{"skillNo":101,"name":102,"nameZh":6,"description":103,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":104,"categories":105,"keywords":107,"publishedAt":35},"01M6F9ZKZJB8H3GWFJ95P6398V","workflow-manager","workflow(org 단위 조립물 — stage DAG 가 워크스페이스의 skill·agent 를 참조)의 대화형 매니저이자 작성 표준의 권위. 의도에서 workflow YAML 을 합성·검토·개선하도록 안내하고, 좋은 워크플로 노하우(본질 원칙=목표·근거·방법→결론·핸드오프 \u002F DAG 무순환 \u002F exit_criteria measurable \u002F 기획 3대 편향 가드)를 보유한다. 결정적 동작(new\u002Flist\u002Fshow\u002Fvalidate)은 solosquad workflow * 헬퍼로 위임. 사용 시점 — 워크플로를 새로 만들·고칠 때, 또는 다른 워크플로의 작성 품질을 판단할 때.","https:\u002F\u002Fgithub.com\u002FAdelie-Squad\u002Fsolosquad\u002Fblob\u002Fmain\u002Fskills\u002Fworkflow-manager\u002FSKILL.md",[106],{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},[81,108,94,109,110,31,96,32,111,112,34,113,114,29,115,79,116,117,118,119],"workflow manager","solosquad","exit","org","make","show","list","yaml","core","validate","dag","new",{"skillNo":121,"name":122,"nameZh":6,"description":123,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":124,"categories":125,"keywords":126,"publishedAt":35},"01M6F9ZH962BA0SABW0DKR0FAF","wbs","마일스톤 → WBS (Work Breakdown Structure) 분해. PM 핵심 산출물. team OKR 의 KR 을 분기→주→일 단위 작업으로 변환. dependency graph + critical path 식별.","https:\u002F\u002Fgithub.com\u002FAdelie-Squad\u002Fsolosquad\u002Fblob\u002Fmain\u002Fskills\u002Fwbs\u002FSKILL.md",[],[122,127,128,29,34,32,129,130,131,132,133,134,135,136,137,81,138,139,79,140],"breakdown","path","critical","decomposition","pm","structure","dependency graph","dependency","okr","work","schedule","team","graph","planning",{"skillNo":142,"name":143,"nameZh":6,"description":144,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":145,"categories":146,"keywords":148,"publishedAt":35},"01M6F9ZE85QYBBX4XMM0W71YN3","triage","Chief 의 과제 분류 도구. 사용자 입력을 4-way 분류 (discussion \u002F workflow \u002F cron \u002F goal) + Educational Nudge (불명확 시 KNOWLEDGE 가이드 선제시).","https:\u002F\u002Fgithub.com\u002FAdelie-Squad\u002Fsolosquad\u002Fblob\u002Fmain\u002Fskills\u002Ftriage\u002FSKILL.md",[147],{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},[143,29,149,116,94,150,34,151,32,152,31,153,154],"cron","knowledge","Educational","classify","goal","nudge",{"skillNo":156,"name":157,"nameZh":6,"description":158,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":159,"categories":160,"keywords":162,"publishedAt":35},"01M6F9Z654Q8647PBSK2WJB7NS","skill-manager","skill(SKILL.md)의 대화형 매니저이자 **작성 표준의 권위(authoring authority)**. 새 skill 생성·검토·개선을 Chief 세션 안에서 안내하고, 좋은 SKILL.md 작성 노하우(description 공식·정량 한도·번들 구조·eval 검증·frontmatter 필드 감사)를 보유한다. 결정적 동작(scaffold\u002Flist\u002Fshow\u002Fvalidate)은 `solosquad skill *` 헬퍼로 위임하고, 의도→초안·품질 검토·개선은 대화로 진행한다. 사용 시점 — skill 을 새로 만들 때, 기존 skill 을 고칠 때, 또는 다른 skill 의 작성 품질을 판단할 때. v1.3.6 — 코드 백엔드 `src\u002Fbot\u002Fskill-manager.ts` 위에 작성 권위를 얹음.","https:\u002F\u002Fgithub.com\u002FAdelie-Squad\u002Fsolosquad\u002Fblob\u002Fmain\u002Fskills\u002Fskill-manager\u002FSKILL.md",[161],{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},[81,163,164,109,119,165,80,34,32,166,31,167,113,114,168,169,29,79,116,170,171,117],"skill manager","bot","authority","authoring","new skill","eval","src","skill authoring","scaffold"]