agentleFS
Sign inStart free
friuns2friuns2/codex-mobile/.agents/skills/github-pr-review-comments/SKILL.md

github-pr-review-comments

Skill931 starsChanged 3 months agoMITView on GitHub

Use when the user asks to review GitHub pull requests, identify concrete risks, and write review comments with `gh`. Focus on actionable findings, not summaries.

1 other project uses a version of this.

Adopted by

Most recently taken up by

  • Other versions1 project uses a version of this.

    Word for word · 1

    1. friuns2/codexUI★ 956100% same

    Edited · 0

    None found.

    Wrote their own version · 0

    None found.

    How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.

  • In this repository4 other files it gives its agents.

The file

Show its rules and text

We split this skill into the rules it gives an agent

16 rules. About 0 are common, 15 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈15

    Distinctive · estimate

    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:

    Use this skill when the user asks to:

    line 10
  2. Distinctive:

    check PRs not opened by the repo owner

    line 13
  3. Very specific:

    write review comments back to GitHub with gh

    line 15
  4. Distinctive:

    Default to a code review mindset:

    line 23
  5. Distinctive:

    avoid low-value style commentary unless it masks a real bug

    line 29
  6. Distinctive:

    If the user asks for PRs not by owner, filter out PRs whose author matches the repo owner.

    line 37
  7. Distinctive:

    Read PR title, author, body, changed files, and commits:

    line 40
  8. Distinctive:

    Prefer 0-3 real findings over a long weak list.

    line 51
  9. Distinctive:

    Write comments when requested

    line 53
  10. Distinctive:

    Keep comments short, direct, and technical.

    line 55
  11. Distinctive:

    Use a terse, high-conviction review style:

    line 60
  12. Distinctive:

    Prefer reviewing non-owner PRs first when the user asks broadly.

    line 83
  13. Distinctive:

    If a finding is platform-related, explicitly call out the platform that should be tested, for example WSL, Linux, macOS, or Windows.

    line 84
  14. Distinctive:

    If you post comments, preserve the user’s requested tone as long as it stays professional and non-abusive.

    line 85
  15. Distinctive:

    When reporting back to the user:

    line 89
  16. Distinctive:

    If no real findings are found, say so directly.

    line 96

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 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

GitHub PR Review Comments

When To Use

Use this skill when the user asks to:

  • review open GitHub PRs
  • check PRs not opened by the repo owner
  • identify bugs, regressions, or missing validation in a PR
  • write review comments back to GitHub with gh

Goal

Produce high-signal PR review findings and, when requested, post concise review comments with gh.

Review Standard

Default to a code review mindset:

  • prioritize correctness bugs
  • prioritize behavioral regressions
  • prioritize compatibility and rollout risk
  • prioritize missing validation where the blast radius is meaningful
  • avoid low-value style commentary unless it masks a real bug

A review comment should only be posted if it says something concrete that could block, regress, or materially weaken the change.

Workflow

  1. Identify candidate PRs
  • Prefer gh pr list --repo friuns2/codexUI --state open.
  • If the user asks for PRs not by owner, filter out PRs whose author matches the repo owner.
  1. Gather PR context
  • Read PR title, author, body, changed files, and commits:
    • gh pr view <number> --repo friuns2/codexUI --json ...
  • Read the diff:
    • gh pr diff <number> --repo friuns2/codexUI
  • Narrow attention to risky files and changed contracts.
  1. Form findings
  • Each finding should answer:
    • what can break
    • why it can break
    • where in the patch it comes from
  • Prefer 0-3 real findings over a long weak list.
  1. Write comments when requested
  • Use gh pr comment <number> --repo friuns2/codexUI --body ...
  • Keep comments short, direct, and technical.
  • One comment per distinct issue is preferred when the issues are unrelated.

Comment Style

Use a terse, high-conviction review style:

  • direct
  • unsentimental
  • technical
  • specific about the failure mode

Good pattern:

  1. Name the concern in one sentence.
  2. Explain the concrete breakage path.
  3. State what needs to be clarified, tested, or fixed before merge.

Avoid:

  • excessive hedging
  • praise padding
  • long summaries of the PR before the issue
  • vague “maybe consider” language when the risk is real

Repo-Specific Notes

  • Canonical GitHub repo: friuns2/codexUI
  • Prefer reviewing non-owner PRs first when the user asks broadly.
  • If a finding is platform-related, explicitly call out the platform that should be tested, for example WSL, Linux, macOS, or Windows.
  • If you post comments, preserve the user’s requested tone as long as it stays professional and non-abusive.

Output Expectations

When reporting back to the user:

  • list findings first
  • include PR number
  • include the short reason
  • mention whether comments were posted

If no real findings are found, say so directly.

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.

Check your setup, free