agentsclimarketplace

Review pr

Skill HsuanYuLee/polaris/.claude/skills/review-pr

Review someone else's PR as a code reviewer: read the PR diff, check against .claude/rules, leave inline comments on issues found, and submit a review with APPROVE or REQUEST_CHANGES. Use when the user asks the assistant to review a PR (subject omitted or = self), e.g.: "review PR", "review 這個 PR", "review 此 PR", "review 該 PR", "幫我 review 這個 PR" (without team subject), "review for me", "code review", or shares a PR URL with self-directed review intent. NOT for "請<同仁/大家/人名>幫我 review" (subject = others) — that is 催 review, route to check-pr-approvals. NOT for "review 大家的 PR" / "掃 PR" (object = others' PRs) — route to review-inbox. NOT for fixing review comments on your own PR — route to engineering revision mode. (use engineering revision mode for that).From its SKILL.md

Install
npx -y skills add HsuanYuLee/polaris --skill review-pr

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

2 things to look at

  • 5 stars5 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
  • runs commandsInstructs the agent to run 1 command, including `post-task-reflection-checkpoint.md`.

SKILL.md

4.3 KB, 990 tokens by cl100k_base, as published. Nobody here has run it

review-pr

以 reviewer 角色審查別人的單一 PR,依 repo rules / handbook / diff context 留 inline comments,並提交 GitHub review。

Contract

此 skill 只處理單一 PR review。多 PR discovery 與 batch orchestration 交給 review-inbox;修自己的 PR review comments 交給 engineering revision mode。

Reviewer stance:prioritize bugs、behavior regressions、security、type safety、project rule violations、missing tests。不要用 personal style preference 擋 merge。 reviewer-side 屬 read-only lane,但對 changes_requestedactive unresolved commentsawaiting_re_reviewmergeable_ready 的語義必須與 author-side mutable lane 一致;不得自行重寫。 本 skill 可輸出 reviewer 結論(APPROVE / COMMENT / REQUEST_CHANGES),但不得把 reviewer 結論寫成 author-side stage authority;「可 merge / 可 release / 已完成」仍必須沿用 shared PR / workflow state,而不是 reviewer prose。

Reference Loading

SituationLoad
Any runreview-pr-entry-fetch-flow.md, pr-input-resolver.md, workspace-config-reader.md, shared-defaults.md
Analysisreview-pr-analysis-flow.md, repo-handbook.md, library-change-protocol.md as needed
Submit and notifyreview-pr-submit-flow.md, workspace-language-policy.md, external-write-gate.md, github-slack-user-mapping.md
Re-reviewreview-pr-rereview-learning-flow.md, review-lesson-extraction.md, repo-handbook.md

Large PR 分批 review 可派 sub-agent;所有 dispatch 必須注入 sub-agent-roles.md 的 Completion Envelope。Sub-agent 只做 analysis,不提交 review、不改檔。

Flow

  1. 從使用者輸入或 Slack context 解析 PR URL;找不到單一 PR 時停止或轉 review-inbox
  2. pr-input-resolver.md 解析 owner、repo、number、本地 project path;找不到本地 repo 時使用 remote read mode。
  3. 用 bundled fetch script 取得 metadata、files、review strategy、existing reviews、approval state、re-review signal。
  4. 讀 repo rules、workspace handbook、PR description、changed files、diff、既有 review comments,建立去重清單。
  5. Review changed files;large PR 依 reference 分組派 sub-agent fan-out。
  6. 合併 findings,依 severity 決定 APPROVECOMMENT、或 REQUEST_CHANGES
  7. Review body、inline comments、Slack notification 送出前跑 language gate。
  8. Submit GitHub review,查詢 approve status,輸出摘要。
  9. 若有 validated repo-specific pattern,依 standard-first rule 更新 handbook。
  10. Slack source 時回覆原始 thread。

Severity Boundary

must-fix 必須是可從 code / diff / rules 直接證明會造成 bug、安全風險、型別錯誤、 或違反關鍵規範。外部 API 行為、language/library behavior、或僅基於慣例的推論,在未驗證前 最多是 should-fix

Write Rules

  • GitHub review、inline comments、Slack replies 都是 external write。
  • 使用 workspace-language-policy.md 或 external write gate 驗證 final text。
  • 不重複留言已由其他 reviewer 指出的同語意問題。
  • Suggested change 只在能精準替換 diff range 時使用。

Completion

輸出 PR、review result、must-fix / should-fix / nit counts、approve status、Slack notification status,以及 handbook updates if any。

Post-Task Reflection (required)

Execute post-task-reflection-checkpoint.md before reporting completion.

What ships with it: 28 files

200.7 KB alongside SKILL.md, 18 of them executable

evals/

Keep looking

Skills are one crate of 325,949. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.