원클릭으로
ship
Use when ready to release - verifies all evidence, runs ship gate, handles deployment, triggers compound janitor
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
메뉴
Use when ready to release - verifies all evidence, runs ship gate, handles deployment, triggers compound janitor
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
SOC 직업 분류 기준
Use after implementation to run independent code review - dispatches reviewer agents based on task level, ensures author-reviewer separation
Use for the L2+ browser-QA gate. Real-browser mode (Playwright Chromium) is opt-in (--browse / SGC_QA_REAL=1); by default runs a stub returning concern (never rubber-stamps). Writes verdict + findings to reviews/{task}/qa/.
Use when starting any conversation - establishes SGC commands, routes tasks to appropriate skills, and enforces system invariants
Use to capture knowledge from solved problems - extracts structured solution documents via 4 specialized agents with dedup enforcement
Use when requirements are unclear, before planning - clarifies goals, constraints, and acceptance criteria through structured questioning
Use when starting any non-trivial task - classifies task level (L0-L3), runs appropriate planning agents, produces intent document and execution plan
| name | ship |
| description | Use when ready to release - verifies all evidence, runs ship gate, handles deployment, triggers compound janitor |
Run the ship gate, write immutable decisions/{id}/ship.md, optionally open a PR, auto-trigger janitor.compound. Every gate failure surfaces a specific error — do not bypass.
Core principle: every ship is backed by on-disk evidence. No evidence, no ship.
/ship after /work + /review (+ /qa for L2+)progress/feature-list.md marked done| Directory | Access |
|---|---|
| decisions | R+W |
| progress | R |
| solutions | — |
| reviews | RW |
src/commands/ship.ts (runShip)fail verdicts (without --override ≥40)--auto; prompts interactive yes confirmation (§4)src/dispatcher/gh-runner.ts — --pr shells gh pr createsrc/dispatcher/agents/janitor-compound.ts — decision logged even on skip (§6)When this skill is invoked, dispatch to the sgc CLI:
bun src/sgc.ts ship $ARGUMENTS
Each gate names itself in the error (features not done, no code reviews, qa evidence missing, fail verdict without override). Fix the underlying evidence; do not retry with --auto.
If the janitor decides compound or update_existing, runCompound executes inside ship. Failure there is non-fatal to ship — ship.md is already committed; the compound failure is logged but the ship decision stands. §10 atomicity guarantees no partial solution write.
For git/PR/deploy workflows beyond sgc's ship gate:
gs:/ship — detect base branch, run tests, bump VERSION, update CHANGELOG, create PRgs:/land-and-deploy — merge the PR, wait for CI, verify production health