菜单
Use once plan is locked and implementation needs to start. Guides a disciplined build loop with scope freeze, milestone-sized slices, validation after each slice, and document sync. Replaces the "start coding and see what happens" failure mode with a controlled march through the plan.
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
| name | vibe-coding-workflow |
| description | Use once plan is locked and implementation needs to start. Guides a disciplined build loop with scope freeze, milestone-sized slices, validation after each slice, and document sync. Replaces the "start coding and see what happens" failure mode with a controlled march through the plan. |
Plan 이 잠긴 뒤 구현 단계를 흐트러지지 않게 진행 하는 skill.
runtime.json → gates.scope_locked == true AND gates.plan_reviewed == truePrompt.md / PRD.md / Plan.md 가 준비됨/office-hours, /product-planner, /plan-eng-review)구현 중 자주 터지는 세 가지 실패 모드를 구조적으로 막는다:
- runtime.json → gates.scope_locked: true
- runtime.json → gates.plan_reviewed: true
- Plan.md 의 milestone + slice 순서 확정
- Implement.md 빈 템플릿 준비
하나라도 안 되면 이 skill 대신 /plan-eng-review 로 돌아간다.
각 slice 는 다음 사이클:
write → test → validate → commit → update doc
구현 중 "이거 고치다 보니 다른 파일도..." 가 떠오르면:
/plan-eng-review 로 돌아가서 Plan 갱신runtime_gate.py verify-implementation 으로 implementation_verified=true 전환| 상황 | 행동 |
|---|---|
| Plan 에 있는 범위 안의 버그 발견 | 현재 slice 에 포함, 고침 |
| Plan 에 없는 영향도 발견 | Implement.md 발견 사항 기록, 현재 slice 완료 후 plan 재검토 |
| 테스트가 애매하게 통과 | Gate ④ runner 로그 확인. "어떤 assertion 이 진짜 이 변경을 증명하나?" |
| "이거 refactor 하고 싶다" 유혹 | Plan 에 없으면 금지. Documentation.md 에 "refactor candidate" 기록하고 넘어감 |
| 두 milestone 을 합치면 편할 것 같다 | 병합 금지. 각 milestone 의 독립 검증 가능성이 중요 |
매 slice 완료 시 Implement.md 업데이트:
## Slice N — <날짜>
### Write paths
- <파일>
### Test / validation
- <실행한 명령 + 결과>
### 커밋
- <SHA> <type>(<scope>): <요약>
### 발견 사항 (현재 slice 밖)
- <있으면 기록, 없으면 "none">
### 다음 slice
- <다음 slice 이름 or "milestone 완료 → 다음 milestone">
Prompt.md, PRD.md, Plan.mdtemplates/Implement.md — 양식CLAUDE.md 의 라우팅 정책 (tier × complexity)product-planner / plan-eng-review 의 verdictExplicit GPT-5.6 native-first router for auditing, designing, installing, or validating Codex skill combinations, global skill setup, routing fixtures, and harness behavior. Use when the user explicitly invokes $codex-skill-router or asks to change or inspect Codex skills, routing, AGENTS guidance, or global harness configuration. Do not use for ordinary coding, review, debugging, research, or UI work merely because the task is complex.
Use when work introduces or changes shared nouns, states, IDs, ownership, authority, or customer-facing vocabulary.
Use when changing harness gates, routing rules, skills, or automation behavior and benchmark/eval evidence is needed to avoid slowing normal coding performance. Do not use for ordinary app code changes.
Use when a request is early, ambiguous, or lacks a codebase-backed shared language; interview the user to expose assumptions before planning.
Use when a codebase, docs, ADRs, tests, prior reviews, or team traces exist; align new work with existing shared language before implementation.
Use first for normal or high-risk natural-language work to choose task mode, tier, grill path, spec layers, team-rule scan, and required gates before implementation.