agentleFS
salvo8519salvo8519/fairmark/.claude/skills/pr-triage/SKILL.md

pr-triage

Skill1 starsChanged 2 days agoApache-2.0View on GitHub

Review the queue of open pull requests on Fairmark — CI and mergeability, conventions, linked issue and spec, numerical and contract changes, documentation and release notes — post factual review comments on GitHub and give the maintainer a merge recommendation for each. Use when the maintainer asks to review pending pull requests or the merge queue.

22 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use aeon-update, auto-workflow and 10 more.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • aaronjmarsSkill

    aeon-update

    Pull framework updates from the upstream Aeon repo into this instance - 3-way merges canon's new commits into a PR, never clobbering operator config.

    Also in 4 projects that use this

  • aaronjmarsSkill

    auto-workflow

    Two-mode aeon.yml workflow builder - analyze inspects URLs and emits a tiered, signal-verified skill-enablement plan plus an aeon.yml diff; enable flips slugs to enabled:true and opens a PR.

    Also in 4 projects that use this

  • +10 more from aaronjmars

The file

We split this skill into the rules it gives an agent

3 rules. 0 are common, 2 are distinctive, and 1 is very specific to this project.

  • 0

    Common

    Most of the 19 edited versions we read keep these.

  • 2

    Distinctive

    This author's own choices. Worth a look.

  • 1

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Remove the worktree when done.

    line 35
  2. Very specific:

    Follow the tiers in "Effort proportional to risk" (AGENTS.md): read small pull requests yourself.

    line 38
  3. Distinctive:

    Do not repeat what CI already reports in its own annotations.

    line 54

“Common” is measured: kept by at least 6 in 10 of the 19 edited versions of this file we could read. “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
<!-- Generated from .agents/skills/pr-triage/SKILL.md by scripts/sync_agents.py. Edit the source, not this file. -->

Talk to the maintainer in their preferred language; everything posted to GitHub is in English, factual, courteous and without icons. Never merge, approve or close a pull request: those decisions belong to the maintainer. Never push to a contributor's branch.

1. Inventory

uv run python scripts/project_state.py --json gives open pull requests with checks and reviews. For each pull request also read, with gh api: repos/{owner}/{repo}/pulls/<n> (mergeable state, base, draft, author, body), .../pulls/<n>/files (changed paths) and .../pulls/<n>/commits.

2. Deterministic checks (per pull request)

CheckHow
Conventionspython scripts/conventions.py pr --title ... --branch ... --body ... --label ...
CIall check runs green; name the failing jobs
Up to datemergeable_state is not behind or dirty (conflicts)
Linked issuethe issue exists, is open, is in the current milestone or sprint
Specmethod changes cite an Approved spec (specs/NNNN-*/spec.md)
Golden casesexisting expected values unchanged, or label numerical-change and a Numerical-change: trailer
Contractsschemas/ changes are additive, or label breaking-change
Documentationscripts/docs_sync.py impact --base origin/main on the pull request head
Release notea fragment in changes/, or label no-release-note
Size and scopeone concern; flag unrelated files and diffs above about 600 changed lines
Overlapsother open pull requests touching the same files (merge-order risk)

To run local checks on a pull request, fetch it into a worktree, never into the current branch: git fetch origin pull/<n>/head:pr-<n> && git worktree add ../fairmark-pr-<n> pr-<n>. Remove the worktree when done.

3. Judgement (only for pull requests that pass step 2 or are close)

Follow the tiers in "Effort proportional to risk" (AGENTS.md): read small pull requests yourself; delegate substantial ones to code-reviewer (diff against AGENTS.md, the constitution and the spec); add numerical-auditor only for the numerical-change tier and security-reviewer when dependencies or workflows change. Do not repeat a review already recorded on the pull request for the same head. Bot pull requests (Dependabot) need only security-reviewer for major versions and the release notes of the dependency.

4. Comments

Show the maintainer every comment before posting it, then post one review per pull request: gh api -X POST repos/{owner}/{repo}/pulls/<n>/reviews -f event=COMMENT -f body=@review.md (inline comments: add comments[] with path, line, body). Structure the body as:

  • a one-sentence summary of what the pull request does;
  • Required changes: numbered, each with the file, the rule it breaks (link to docs/en/project/conventions.md, AGENTS.md or the spec) and the fix;
  • Suggestions: optional improvements, clearly marked as optional;
  • a thank-you line for external contributors. Do not repeat what CI already reports in its own annotations; refer to the job instead.

5. Report to the maintainer

One table, ordered by recommended merge sequence:

PRTitleCIConventionsRiskRecommendation

Recommendations: Merge (all green, no open points), Merge after rebase, Changes requested (comment posted), Needs your decision (methodology, breaking change, numerical change), Close (superseded or out of scope; propose the closing comment). For overlapping pull requests, say which to merge first and why.

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.