Menu
Implement the planned change task by task, then spawn the independent verify stage until it passes
Instalar com Codex ou Claude Copie este prompt, cole no Codex, Claude ou outro assistente e deixe que ele revise a página da skill e instale para você.
Baseado na classificação ocupacional SOC
| name | apply |
| description | Implement the planned change task by task, then spawn the independent verify stage until it passes |
| argument-hint | [change-id] |
| effort | max |
gate를 판정한다: applyRequires에 열거된 artifact가 전부 done이어야 한다.
(isComplete는 gate가 아니다 — design은 포함 조건에 해당할 때만 존재한다.)
아니면 구현이 아니라 계획이 부족한 것 — tx:plan으로 돌아간다.
"${CLAUDE_PLUGIN_ROOT}/bin/tx-hook" openspec status --change <change-id> --json
context 파일 경로와 task checklist를 받는다:
"${CLAUDE_PLUGIN_ROOT}/bin/tx-hook" openspec instructions apply --change <change-id> --json
task를 dependency 순서대로 구현한다.
spec wording이 구현을 구속한다 — 구현 중 spec이 틀렸음이 드러나면 tx:plan으로 delta를
고친 뒤 계속한다.
모든 task가 완료되고 change에 spec delta가 있으면 verify stage를 spawn한다 (필수):
Agent(subagent_type="tx:verify", prompt="change-id: <change-id>")
change-id 외에는 아무것도 전달하지 않는다 — verifier는 artifact와 코드를 직접 읽는다. 결함이 보고되면 구현 context가 살아있는 지금 이 자리에서 고치고 재spawn한다 — pass까지.
구현 중 만나는 모든 unknown은 셋 중 하나로 처리한다:
Answer the owner's questions about a running or finished loop from its records — read-only, in a separate session
Hours-long (3-12h) multi-agent code-architecture optimization — cross-examines antipatterns into consensus, then applies only the highest-ROI refactors. Not for routine refactors.
Hours-long (3-12h) multi-agent documentation-architecture optimization — verifies every claim in every non-executable text (markdown, spec systems, comments) against the code, cross-examines discrepancies into consensus, then applies only the highest-ROI fixes. Code is never modified — defects are reported, not fixed. Not for touching up a single file.
Hours-long (3-12h) multi-agent integrity-boundary optimization — hunts every reachable state the boundary (types, invariants, error definitions, tests) fails to contain, then absorbs only the highest-ROI set into it — code and docs both, pinned by tests. Not for fixing a known bug.
Archive the completed change — task gate, delta sync into main specs, revalidate
Close the transaction — verify, archive, docs gate, then squash merge into base after CI passes