メニュー
Run an independent semantic review gate before merging non-trivial work, on top of (not duplicating) Alibaba Code Review and CI. Use after implementation and verification, especially for medium/high-risk changes.
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
Records release intent with Changesets and guides semver, changelog, CI, and artifact-version checks for independently distributed packages. Use when a repository has `@changesets/cli`, `.changeset/`, a version-packages PR, publishable plugins, SDKs, libraries, or monorepos.
Synthesize the current conversation and repository context into a concise product and engineering spec, publish it to the project issue tracker, and gate agent readiness before decomposition. Use when high-level work needs product framing, implementation and testing decisions, or success criteria captured without another general interview.
Add GitHub-visible AI agent and AI code-review metadata and exact agent labels without adding AI attribution to commits, and scope Alibaba/Open Code Review's review surface to control LLM token spend. Use when configuring PR templates, issue/PR labels, PR creation commands, PR update comments, or Alibaba Code Review/open code review workflows to disclose the implementing agent, review tool, LLM model, run URL, or to add/tune a `.opencodereview/rule.json`.
Define acceptance criteria, risk level, and verification before editing. Use at the start of non-trivial implementation or when an issue lacks clear proof of completion.
Check deployment, staging, rollback, and health verification before release. Use for user-facing, infrastructure, CI/CD, integration, or high-risk changes.
Clarify non-trivial engineering work before implementation by resolving ambiguity, terminology, constraints, and the smallest viable path. Use when a request is broad, fuzzy, architectural, cross-cutting, or has hidden product decisions.
| name | review-gate |
| description | Run an independent semantic review gate before merging non-trivial work, on top of (not duplicating) Alibaba Code Review and CI. Use after implementation and verification, especially for medium/high-risk changes. |
Use this after implementation and before merge.
ai-agent-pr-metadata.The reviewer should read the issue, acceptance criteria, Alibaba Code Review's output, CI results, and the PR diff first. Pull more context only when needed.
CI remains a hard merge gate; this review does not replace it and is not the place to rerun or restate lint/static-analysis automation. Do not merge non-trivial work without a review record. If an automated or subagent reviewer stalls, post a manual expert review that states residual risk.