agentleFS
fpindejfpindej/netrock/.claude/skills/address-review/SKILL.md

address-review

Skill234 starsChanged 3 months agoMITView on GitHub

Read PR review comments from GitHub, evaluate, address, and reply

22 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use open-pr, pr-description and 3 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.
  • Google CloudSkill

    open-pr

    Commits local changes to a new Git branch and opens a GitHub pull request through the configured GitHub MCP server. Use when proposing code changes in a new pull request. Don't use for local-only commits without a remote GitHub repository.

    Also in 3 projects that use this

  • MicrosoftSkill

    pr-description

    Draft Nacelle PR titles, pull request descriptions, commit titles, commit descriptions, and squash commit messages. Use when asked for PR or commit wording, a branch summary, or a PR description that can be used in a commit. Omit validation sections from commit bodies and remove filler.

    Also in 3 projects that use this

  • MicrosoftSkill

    pr-review

    Verify the technical accuracy of a vscode-docs pull request against the VS Code source code in microsoft/vscode. Use when reviewing a docs PR for factual correctness: setting names, command IDs, default values, API shapes, keybindings, versioned availability, and described behavior.

    Also in 3 projects that use this

  • Hugging FaceSkill

    implement

    Make tests pass. Invoke after /write-tests produces failing tests.

    Also in 3 projects that use this

  • +1 more from microsoft

The file

Show its rules and text

We split this skill into the rules it gives an agent

13 rules. 0 are common, 12 are distinctive, and 1 is very specific to this project.

  • 0

    Common

    Most of the 22 edited versions we read keep these.

  • 12

    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:

    Read review comments on a GitHub PR, evaluate each one, and address it appropriately.

    line 9
  2. Distinctive:

    Resolve the PR number: If an argument is provided, use it.

    line 13
  3. Distinctive:

    Read the referenced file + surrounding context

    line 34
  4. Distinctive:

    If yes: make the change, reply with what was changed and which commit

    line 36
  5. Distinctive:

    If the change would break something or is incorrect: explain why and propose an alternative

    line 37
  6. Distinctive:

    Verify the claim by reading the actual code/schema.

    line 53
  7. Distinctive:

    Do NOT resolve threads that need further discussion from the reviewer.

    line 61
  8. Distinctive:

    Read the FULL thread before responding (earlier replies give context)

    line 69
  9. Distinctive:

    If a change requires a migration, new endpoint, or significant scope: discuss before implementing

    line 71
  10. Distinctive:

    Never resolve a thread without either fixing the code or explaining why no change is needed

    line 72
  11. Distinctive:

    If review comments contradict each other, flag the conflict for the user instead of trying to satisfy each of them

    line 73
  12. Very specific:

    Run /verify before pushing

    line 75
  13. Distinctive:

    No em dashes in replies

    line 76

“Common” is measured: kept by at least 6 in 10 of the 22 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

/address-review

Read review comments on a GitHub PR, evaluate each one, and address it appropriately.

Steps

  1. Resolve the PR number. If an argument is provided, use it. Otherwise, detect from current branch:
gh pr view --json number,title,url
  1. Fetch all review comments (includes inline code comments and review threads):
gh api --method GET repos/{owner}/{repo}/pulls/{number}/comments
gh api --method GET repos/{owner}/{repo}/pulls/{number}/reviews

(Always pass --method GET explicitly for gh api reads - only explicit-GET calls are pre-approved by the project permissions.)

Filter to only unresolved threads with actual review content. Read the full thread (all replies in each discussion), not just the first comment.

  1. For each unresolved thread, evaluate what type of comment it is:

A. Change request ("change this", "use X instead of Y", "add Z")

  1. Read the referenced file + surrounding context
  2. Evaluate if the change is correct, factual, and worth it
  3. If yes: make the change, reply with what was changed and which commit
  4. If the change would break something or is incorrect: explain why and propose an alternative

B. Question ("why is this here?", "doesn't X already handle...?")

Reply with a clear explanation. Reference specific code/docs. If the answer reveals a real issue, fix it too.

C. Suggestion / discussion ("we could...", "what if we...")

Evaluate effort vs impact. Reply with one of:

  • "Good idea, I'll do it in this PR" (and do it)
  • "Makes sense, but better as a follow-up" (explain why)
  • "I don't think it's needed because..." (explain reasoning)

D. Factual correction ("that's not true", "table doesn't have column X")

Verify the claim by reading the actual code/schema. If the reviewer is right, fix it. If wrong, explain with evidence.

  1. Reply to each thread:
gh api repos/{owner}/{repo}/pulls/{number}/comments/{comment_id}/replies --method POST -f "body=<reply>"

Do NOT resolve threads that need further discussion from the reviewer.

  1. After addressing all comments, run /verify.
  2. Commit and push with a descriptive message like fix: address review - <summary of changes>.

Rules

  • Read the FULL thread before responding (earlier replies give context)
  • Reference specific commits when you make a change
  • If a change requires a migration, new endpoint, or significant scope: discuss before implementing
  • Never resolve a thread without either fixing the code or explaining why no change is needed
  • If review comments contradict each other, flag the conflict for the user instead of trying to satisfy each of them
  • Group related fixes into a single commit when they address the same concern
  • Run /verify before pushing
  • No em dashes in replies

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.