agentleFS
metyatechmetyatech/skill-pr-review-workflow/SKILL.md

pr-review-workflow

Skill0 starsChanged 4 months agoMITView on GitHub

Use when addressing PR review feedback, re-requesting reviews, handling GitHub review bot re-reviews (Codex, Copilot), or managing GitHub notifications. Do not use for creating new PRs or general development.

The file

We split this skill into the rules it gives an agent

7 rules. About 0 are common, 7 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈7

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    After addressing PR review feedback, the agent MUST resolve the corresponding review threads before concluding.

    line 10
  2. Distinctive:

    Before re-requesting review, the agent MUST run the relevant verification suite and MUST summarize results (commands and outcomes) in the …

    line 13
  3. Distinctive:

    After pushing fixes, the agent MUST re-request review only from reviewers who posted addressed feedback in the current round.

    line 26
  4. Distinctive:

    If no applicable reviewer remains, the agent MUST ask who should review next.

    line 31
  5. Distinctive:

    When Codex or Copilot review bots are configured for the repository, the agent MUST trigger re-review only for the bot(s) that posted …

    line 36
  6. Distinctive:

    After processing notifications, the agent MUST bulk-delete any remaining read-but-not-done notifications using the same DELETE API.

    line 61
  7. Distinctive:

    After completing a PR, the agent MUST merge it, sync the target branch, and delete the PR branch locally and remotely.

    line 67

≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

PR review feedback workflow

Addressing review feedback

  • After addressing PR review feedback, the agent MUST resolve the corresponding review threads before concluding. If the agent lacks permission, the agent MUST state it explicitly.
  • Before re-requesting review, the agent MUST run the relevant verification suite and MUST summarize results (commands and outcomes) in the PR comment or description.
  • For feedback caused by an agent-originated miss or a reusable reasoning gap, the agent MUST assess whether recurrence prevention belongs in tests, checks, skills, or rules. When a rule or skill update is warranted, the agent MUST make the update as part of the same change set. When it is not warranted, the agent MUST briefly state why the earliest reusable gate is elsewhere.

Re-requesting reviews

  • After pushing fixes, the agent MUST re-request review only from reviewers who posted addressed feedback in the current round.
  • The agent MUST NOT re-request review from reviewers (including AI reviewers) who did not post addressed feedback or who already indicated no actionable issues.
  • If no applicable reviewer remains, the agent MUST ask who should review next.

AI review bot re-review

When Codex or Copilot review bots are configured for the repository, the agent MUST trigger re-review only for the bot(s) that posted addressed feedback.

  • For Codex re-review: comment @codex review on the PR.
  • For Copilot re-review: use the GitHub API to remove and re-request the bot reviewer copilot-pull-request-reviewer[bot]. The agent MUST NOT rely on UI-based reviewer assignment.
    • Remove: DELETE /repos/{owner}/{repo}/pulls/{pr}/requested_reviewers with body {"reviewers":["copilot-pull-request-reviewer[bot]"]}.
    • Add: POST /repos/{owner}/{repo}/pulls/{pr}/requested_reviewers with body {"reviewers":["copilot-pull-request-reviewer[bot]"]}.

GitHub notification management

  • The agent MUST use DELETE /notifications/threads/{id} (HTTP 204) to mark notifications as done (removes the notification from the inbox and moves it to the Done tab).
  • The agent MUST NOT use PATCH /notifications/threads/{id}, which marks as read but leaves the notification in the inbox.
  • After processing notifications, the agent MUST bulk-delete any remaining read-but-not-done notifications using the same DELETE API.

Post-merge cleanup

  • After completing a PR, the agent MUST merge it, sync the target branch, and delete the PR branch locally and remotely.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.