ワンクリックで
guide
Codex Stage 1 guide for review-loop and the shared Claude/Codex state model.
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。
メニュー
Codex Stage 1 guide for review-loop and the shared Claude/Codex state model.
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
Codex Stage 1 execute skill. Run the execution + quality polish + delivery stages of review-loop against one of three entry modes: resume an approved session (`--session`), execute a user-supplied plan (`--plan`), or run a pure-CR pass on the current working tree (`--review-only`). Supports batched runs via `--stop-after <stage>`. Use when you already have a plan, or only want CR on existing code.
Codex-native Stage 1 review-loop skill. Orchestrates planning and execution with a Codex Executor and Claude/Codex reviewer backends while sharing the .review-loop protocol with Claude Code.
Run the execution + quality polish + delivery stages of review-loop against one of three entry modes: resume an approved session (`--session`), execute a user-supplied plan (`--plan`), or run a pure-CR pass on the current working tree (`--review-only`). Supports batched runs via `--stop-after <stage>`. Use when you already have a plan, or only want CR on existing code.
Automates a Plan-Execute-Review workflow with built-in iteration loops. An Orchestrator coordinates an Executor sub-agent and a configurable Reviewer (external AI CLI like Codex, or Claude sub-agent) to drive work items from description to delivery — with every review finding visible to the user. Trigger: "run review-loop on", "start review loop", "let the agents handle", or any task the user wants driven by a plan→review→implement→CR cycle.
Codex Stage 1 planning-only skill. Drives a work item from raw description to a reviewer-approved plan in `.review-loop/sessions/{uuid}.md`, then exits with a hint to resume via `review-loop:execute --session <uuid>`. Use when you want plan-only iteration without immediately entering execution.
Run the planning phase only: drive a work item from raw description to a reviewer-approved plan in `.review-loop/sessions/{uuid}.md`, then exit with a hint to resume via `review-loop:execute --session <uuid>`. Use when you want plan-only iteration without immediately entering execution.
| name | guide |
| description | Codex Stage 1 guide for review-loop and the shared Claude/Codex state model. |
review-loop Does in CodexIn Codex Stage 1, review-loop is a repo skill that follows the same broad
review workflow used in Claude Code. It coordinates planning, implementation,
and review while keeping the shared session log current.
Codex Stage 1 follows the same broad exec -> polish -> docs -> security -> delivery lifecycle.
Codex reads and writes the same review-loop state as Claude Code:
.review-loop/config.md.review-loop/sessions/That means a project can keep one shared config file and one shared session log history across both runtimes. Codex Stage 1 assumes a single orchestrator-owned workspace for the session.
Stage 1 in Codex includes only:
review-loopguideIt does not yet migrate:
code-quality-loopreview-prreorganizeCodex Stage 1 defaults to the outside-sandbox Claude CLI reviewer path. In
practice, that means review stays on claude -p --model ... unless the user
explicitly opts into the local Codex reviewer.
You can force the local Codex reviewer with:
codex_reviewer_backend: codexThis is the override to use when you want Codex to skip the Claude CLI reviewer
and use the Codex reviewer directly. In that case, codex_reviewer_model is
the paired model override, while reviewer_model still applies to the Claude
CLI reviewer path and judgment_model is its shared-tier fallback before the
explicit claude-sonnet-4-6 backstop.
cheap_model is accepted in the shared config so Claude and Codex can share
the same file, but in Codex Stage 1 it is a documented no-op because only
judgment-tier Codex agents are currently shipped.
quality_focus applies only when Step 3.5 Quality Polish actually runs.
skip_quality_polish: true mints polish as a no-op completion and still continues through docs and security.
before-polish, before-docs, and before-security as clean stop points..agents/skills/ in the Codex workspace..review-loop/config.md..review-loop/sessions/.