agentleFS
Sign inStart free
mnapolimnapoli/skills/address-pr-review/SKILL.md

address-pr-review

Skill21 starsChanged 41 days agoMITView on GitHub

Look at the pull request review comments and address any issues raised.

7 other projects use a version of this, 1 of them edited.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

12 rules. 0 are common, 9 are distinctive, and 3 are very specific to this project.

  • 0

    Common

    Most of the 6 edited versions we read keep these.

  • 9

    Distinctive

    This author's own choices. Worth a look.

  • 3

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Run both helper scripts located in this skill's directory (the directory containing this SKILL.md):

    line 11
  2. Distinctive:

    If there are no unresolved threads and CI is green, stop here.

    line 30
  3. Distinctive:

    If there are many items, create one task per item to keep things organized.

    line 34
  4. Distinctive:

    Read the comment thread to understand the issue

    line 38
  5. Distinctive:

    Treat the comment as a challenge, not an order.

    line 39
  6. Distinctive:

    If a change is needed, make the necessary code changes

    line 41
  7. Distinctive:

    Read the failure output to understand what broke

    line 45
  8. Distinctive:

    After addressing all review comments and CI failures, commit your changes and push.

    line 50
  9. Distinctive:

    After addressing each comment, reply using the helper script:

    line 54
  10. Very specific:

    If the code was fixed as mentioned in the thread, reply with simply: "Fixed as suggested." and pass --resolve to mark the thread as …

    line 60
  11. Very specific:

    If no change was applied, briefly explain why and keep the thread unresolved (omit --resolve) so the reviewer can reply if they disagree.

    line 61
  12. Distinctive:

    Don't be unnecessarily verbose.

    line 62

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

First, run git branch --show-current to retrieve the current branch. Do not trust branch/git status information present in the conversation context as it may be stale.

Step 1: Fetch unresolved review threads and CI status

Run both helper scripts located in this skill's directory (the directory containing this SKILL.md):

bash <skill-directory>/get-unresolved-threads.sh
bash <skill-directory>/get-failing-ci.sh

The first returns a JSON array of unresolved threads. Each thread has:

  • id — thread ID (needed to reply)
  • path — file path
  • line / startLine — line numbers in the diff
  • isOutdated — whether the diff has changed since the comment
  • comments.nodes[] — the comment thread (first = original comment, rest = replies), each with author.login, body, and url

The second checks CI status and outputs failure details if any jobs failed (exit code 1 = failures, 0 = all passed).

If there are no unresolved threads and CI is green, stop here.

Step 2: Address each issue

If there are many items, create one task per item to keep things organized.

For each unresolved review thread:

  1. Read the comment thread to understand the issue
  2. Treat the comment as a challenge, not an order. The reviewer may lack context: the comment can conflict with a decision made earlier during the implementation, or ask for a refactoring out of proportion with the issue.
  3. Determine if a code change is needed, or if no change is needed (e.g. the comment is invalid or the suggestion is not appropriate)
  4. If a change is needed, make the necessary code changes

For each CI failure:

  1. Read the failure output to understand what broke
  2. Fix the code or tests to make CI pass

Step 3: Commit and push changes

After addressing all review comments and CI failures, commit your changes and push.

Step 4: Reply to each review thread

After addressing each comment, reply using the helper script:

bash <skill-directory>/reply-to-thread.sh "<thread_id>" "<body>" [--resolve]
  • If the code was fixed as mentioned in the thread, reply with simply: "Fixed as suggested." and pass --resolve to mark the thread as resolved.
  • If no change was applied, briefly explain why and keep the thread unresolved (omit --resolve) so the reviewer can reply if they disagree.
  • Don't be unnecessarily verbose.

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