agentleFS
lgtm-hqlgtm-hq/ai-skills/skills/pr/SKILL.md

pr

Skill1 starsChanged 10 days agoMITView on GitHub

Create pull requests with proper templates and metadata. Use when asked to create a PR, open a pull request, or submit changes for review. Auto-assign and auto-labeling handled by CI.

  • Commits and pushes

The file

Show its rules and text

We split this skill into the rules it gives an agent

17 rules. About 0 are common, 10 are distinctive, and 7 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈10

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 7

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Create pull requests with proper templates and metadata.

    line 8
  2. Very specific:

    If the branch name contains an issue number (e.g., feat/123-add-feature, fix/45-parser-crash), extract it and fetch the issue details:

    line 18
  3. Distinctive:

    Use the issue context to:

    line 25
  4. Distinctive:

    If no issue number is found in the branch name, skip this step.

    line 30
  5. Very specific:

    Always: use this exact format when linking issues in a PR body.

    line 34
  6. Very specific:

    Use a ## Closes heading (not ## Related Issues)

    line 51
  7. Very specific:

    Use Relates to instead of Closes for issues that are related but not fully resolved by this PR

    line 53
  8. Distinctive:

    Do not add any text after the issue number (no titles, no descriptions)

    line 55
  9. Distinctive:

    When asked to create a PR:

    line 59
  10. Very specific:

    Read the repo's PR template: look for .github/PULL_REQUEST_TEMPLATE.md

    line 64
  11. Distinctive:

    Type: Check the matching type checkbox (feat, fix, test, etc.)

    line 67
  12. Distinctive:

    Testing: Check boxes for tests that were added/run

    line 71
  13. Very specific:

    Quality Checks: Check boxes for lint/format/build that passed

    line 72
  14. Distinctive:

    Checklist: Check all applicable items

    line 73
  15. Distinctive:

    Leave sections like Breaking Changes, Screenshots, Deployment Notes with their placeholder comments if not applicable

    line 74
  16. Distinctive:

    Use a HEREDOC for the body to preserve formatting.

    line 82
  17. Very specific:

    Important: Do NOT use gh pr create --fill — it only uses the commit message and skips the PR template entirely.

    line 84

≈ “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

Pull Request

Create pull requests with proper templates and metadata.

Requirements

  • PR should use the PR template of the current repository
  • Assignee is automatically set by CI (random CODEOWNER)
  • Labels are automatically applied by CI based on changed files

Issue Detection

If the branch name contains an issue number (e.g., feat/123-add-feature, fix/45-parser-crash), extract it and fetch the issue details:

gh issue view <number> --json title,body,labels

Use the issue context to:

  1. Link the issue using the Closes section format (see below)
  2. Inform the PR summary and description from the issue's title and body

If no issue number is found in the branch name, skip this step.

Issue Link Format

Always use this exact format when linking issues in a PR body. Replace the repo's ## Related Issues template section with a ## Closes section:

## Closes

- Closes #123
- Closes #456
- Relates to #789

This format ensures GitHub renders rich issue links (with title, status, etc.) automatically. Do not append the issue title after the number — GitHub handles that via its rich formatting.

Rules:

  • Use a ## Closes heading (not ## Related Issues)
  • Each link is a bullet point: - Closes #<number>
  • Use Relates to instead of Closes for issues that are related but not fully resolved by this PR
  • Do not add any text after the issue number (no titles, no descriptions)

Usage

When asked to create a PR:

  1. Ensure pre-commit checks pass (lint, test, Docker if applicable)
  2. Push the branch to remote
  3. Detect associated issue from branch name (see above)
  4. Read the repo's PR template: look for .github/PULL_REQUEST_TEMPLATE.md
  5. Fill in the PR template with actual content:
    • Summary: Concise description of what the PR does and why
    • Type: Check the matching type checkbox (feat, fix, test, etc.)
    • Changes Made: Bullet list of specific changes
    • Closes: Replace ## Related Issues with ## Closes section using the issue link format specified above
    • Testing: Check boxes for tests that were added/run
    • Quality Checks: Check boxes for lint/format/build that passed
    • Checklist: Check all applicable items
    • Leave sections like Breaking Changes, Screenshots, Deployment Notes with their placeholder comments if not applicable
  6. Create the PR:

    gh pr create --title "<title>" --body "<filled template>"

    Use a HEREDOC for the body to preserve formatting.

Important: Do NOT use gh pr create --fill — it only uses the commit message and skips the PR template entirely. Always read and fill the template.

Note: Assignees and labels are handled automatically by GitHub Actions:

  • pr-auto-assign.yml - Assigns a random CODEOWNER
  • pr-labeler.yml - Labels based on changed files (see .github/labeler.yml)

Checklist

Before creating PR:

  • All linting passes
  • All tests pass
  • Branch is pushed to remote
  • Associated issue detected and linked (if branch has issue number)
  • PR template read and filled with actual content
  • PR created with filled template (not --fill)

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.