agentleFS
quarkiversequarkiverse/quarkus-langchain4j/.agents/skills/pull-requests/SKILL.md

pull-requests

Skill325 starsChanged 7 days agoApache-2.0View on GitHub

Rules for preparing pull requests and commits in the Quarkus project: title conventions, description format, commit hygiene, labels, and contribution policies.

16 other projects use a version of this; 1 new project in the last 4 weeks, against 0 the 4 before.

Adopted by

Surging: 1 new project took it in the last 4 weeks, none the 4 weeks before.thin data

Week of Sep 20: 0 new projectsWeek of Sep 27: 0 new projectsWeek of Oct 4: 0 new projectsWeek of Oct 11: 0 new projectsWeek of Oct 18: 0 new projectsWeek of Oct 25: 0 new projectsWeek of Nov 1: 0 new projectsWeek of Nov 8: 1 new project
Sep 2026Oct 2026Nov 2026
New projects taking a version of this file each week, over the 8 weeks our latest run read: only 1 in all, so this is a hint, not a trend.

Most recently taken up by

The file

Show its rules and text

We split this skill into the rules it gives an agent

23 rules. 0 are common, 13 are distinctive, and 10 are very specific to this project.

  • 0

    Common

    Most of the 15 edited versions we read keep these.

  • 13

    Distinctive

    This author's own choices. Worth a look.

  • 10

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Write a clear, natural-language sentence that describes the change:

    line 13
  2. Very specific:

    Do not prefix with feat:, fix:, chore:, docs:, refactor:, or any other Conventional Commits type — the Quarkus bot will flag this

    line 16
  3. Distinctive:

    Keep it concise but descriptive

    line 18
  4. Distinctive:

    Do not end with a period

    line 19
  5. Distinctive:

    Keep it concise — a short paragraph is usually enough

    line 33
  6. Distinctive:

    Do not include a "Test plan" or "Summary" section

    line 34
  7. Distinctive:

    Do not include "Generated with Claude Code" or similar footers

    line 35
  8. Very specific:

    Do not include Co-Authored-By trailers referencing an AI tool

    line 36
  9. Very specific:

    If your PR introduces a noteworthy feature, include a single line at the end of the description prefixed with Release note: that can be …

    line 40
  10. Distinctive:

    If the change is breaking, explain in the description:

    line 53
  11. Very specific:

    Every pull request fixes or implements an issue.

    line 60
  12. Very specific:

    Write a reference as a bare #1234 so GitHub shows its title.

    line 63
  13. Distinctive:

    Do not open a second pull request with another solution to the same issue.

    line 69
  14. Distinctive:

    When a pull request needs another one first, say so in the description and link it.

    line 72
  15. Distinctive:

    Keep the title in sync with the change.

    line 83
  16. Distinctive:

    Write commit messages as clear, natural-language sentences starting with an uppercase letter — same style as the PR title

    line 91
  17. Very specific:

    Do not add Co-Authored-By trailers referencing AI tools (e.g.

    line 93
  18. Very specific:

    Run ./mvnw process-sources on changed modules to fix formatting

    line 107
  19. Very specific:

    Run tests in Java mode (./mvnw verify -f extensions/<name>/)

    line 108
  20. Very specific:

    If adding a new integration test module, register it in .github/native-tests.json

    line 110
  21. Very specific:

    Update documentation in docs/src/main/asciidoc/ when the change affects user-facing behavior, config, or APIs

    line 112
  22. Distinctive:

    Do not submit code, tests, or descriptions copied directly from an LLM without human review

    line 118
  23. Distinctive:

    Do not use bots or agents to open PRs without human authorship and responsibility

    line 121

“Common” is measured: kept by at least 6 in 10 of the 15 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

Pull Requests

PR Title

Write a clear, natural-language sentence that describes the change:

  • Start with an uppercase letter
  • Do not prefix with feat:, fix:, chore:, docs:, refactor:, or any other Conventional Commits type — the Quarkus bot will flag this
  • Keep it concise but descriptive
  • Do not end with a period

Examples

Add multipart upload support to RESTEasy Reactive
Fix NPE when config property is unset
Update OIDC provider configuration guide
Simplify bean resolution logic in ArC

PR Description

  • Explain why the change is needed, not what changed (the diff shows that)
  • Keep it concise — a short paragraph is usually enough
  • Do not include a "Test plan" or "Summary" section
  • Do not include "Generated with Claude Code" or similar footers
  • Do not include Co-Authored-By trailers referencing an AI tool

Noteworthy features

If your PR introduces a noteworthy feature, include a single line at the end of the description prefixed with Release note: that can be used verbatim in the release notes. Keep it user-facing — focus on the benefit, not the implementation:

Release note: RESTEasy Reactive now supports multipart file uploads with
progress tracking via the new `@PartProgress` annotation.

Only do this for noteworthy features, not small enhancements or bug fixes

Breaking changes

If the change is breaking, explain in the description:

  1. What breaks
  2. Why it was necessary
  3. How users should migrate

Linked issue

Every pull request fixes or implements an issue. When none exists, open it first with the matching template. End the description with Fixes #1234 on its own line. Do not put the issue number in the title. Write a reference as a bare #1234 so GitHub shows its title.

One concern per pull request

  • A pull request fixes one issue or adds one feature. Put an unrelated fix or a clean-up in a separate pull request, even when it is small.
  • Do not open a second pull request with another solution to the same issue. Describe the other solutions in the issue or in the description, and let the reviewers choose.
  • When a pull request needs another one first, say so in the description and link it.

Updating a pull request after review

  • Rebase on main and push with --force-with-lease. Never merge main into the branch: a merge commit fails the "Merge commits" check.
  • Answer every review thread after the push. Start with the answer, then link the commit that applies it.
  • Change the description when the change changed. A reviewer reads the description first, and a stale one costs a review round.
  • Keep the title in sync with the change.

Commits

  • Commits must be atomic and semantic — each commit should represent one logical change
  • Properly squash fixup commits before the PR is ready for merge. Fixup commits are acceptable during review, but must be squashed at the end
  • Write commit messages as clear, natural-language sentences starting with an uppercase letter — same style as the PR title
  • Do not add Co-Authored-By trailers referencing AI tools (e.g. Co-Authored-By: Claude, Co-Authored-By: Copilot)

Labels

Apply these labels when relevant:

LabelWhen to apply
release/noteworthy-featureThe PR introduces a significant user-facing feature that should appear in release notes
release/breaking-changeThe PR contains a breaking change (API removal, behavior change, config rename)

Before Submitting

  1. Run ./mvnw process-sources on changed modules to fix formatting
  2. Run tests in Java mode (./mvnw verify -f extensions/<name>/)
  3. For non-trivial changes, test in native mode (-Dnative)
  4. If adding a new integration test module, register it in .github/native-tests.json
  5. Update documentation in docs/src/main/asciidoc/ when the change affects user-facing behavior, config, or APIs

LLM / Agent Policy

  • You are responsible for validating every change you submit — AI-assisted or not
  • Do not submit code, tests, or descriptions copied directly from an LLM without human review
  • AI-generated tests must validate actual behavior, not just exercise code paths
  • Do not use bots or agents to open PRs without human authorship and responsibility

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.