メニュー
Use when a task needs a bounded retry loop between execution, review, testing, and follow-up fixes
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
SOC 職業分類に基づく
Use only when the user explicitly allows bounded unattended execution of an approved Alloy task.
Use when encountering a bug, regression, failing command, or unexpected behavior before attempting a fix.
Use before planning when requirements are ambiguous, large, user-visible, or likely to cause rework unless decisions are locked.
Use when implementing an approved Alloy plan. Maintains progress.md, enforces TDD/debug discipline, and stops on blocked gates.
Use when a task needs broad source understanding before planning or execution.
Use when turning context or a clear user request into an implementation-ready plan with acceptance criteria, files, commands, gates, risks, and handoff.
| name | ralph-loop |
| description | Use when a task needs a bounded retry loop between execution, review, testing, and follow-up fixes |
Ralph Loop runs recurring work as a bounded retry loop. Use it when an implementation needs repeated executor, reviewer, tester, or verifier passes, but keep the loop finite: define the goal, the stopping condition, the maximum iteration count, and the evidence required before another pass starts.
Start the loop with /ralph-loop and include the task, the reviewers or checks to run, and the maximum number of iterations. Each pass should record what changed, what was verified, and whether the next pass is still justified by concrete findings instead of momentum.
Cancel the active loop with /cancel-ralph. Cancellation should leave all current artifacts intact, summarize the latest completed iteration, and return control to the user without starting another executor pass.
