pr
Skill67 starsChanged 46 days agoApache-2.0View on GitHub
Creates draft pull requests from the project PR template with Conventional Commits titles and an AI attribution footer. Use when creating or opening pull requests, running /pr, pushing a branch for review, or preparing changes for merge. Make sure to use this skill whenever a pull request needs to be created — it is the project's PR workflow and replaces the default one.
- Commits and pushes
In this repository10 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- bdd-test-runnerSkill
- codegenSkill
- commitSkill
- compilationSkill
- feature-optionSkill
- issue-creatorSkill
- kotlin-api-consultantSkill
- sample-scaffolderSkill
- skill-creatorSkill
The file
We split this skill into the rules it gives an agent
4 rules. About 0 are common, 3 are distinctive, and 1 is very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈3
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
- Distinctive: line 34
No commits ahead of main
- Very specific: line 48
Determine scope from paths: use the scope table in the commit skill (compiler, codegen, gradle, annotations, samples, claude, ci).
- Distinctive: line 70
When Claude wrote or co-wrote the change, end the description with an attribution line after the template:
- Distinctive: line 76
If the session already supplies attribution lines for pull requests, use those verbatim instead.
≈ “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.
Files
PR
Creates pull requests — always as drafts, from the project template, crediting AI co-authorship.
Instructions
1. Gather Repository State
# Current branch
git branch --show-current
# Verify not on main/master
git branch --show-current | grep -E '^(main|master)$' && echo "ERROR: Cannot PR from main"
# Check remote tracking
git rev-parse --abbrev-ref --symbolic-full-name @{u} 2>/dev/null || echo "No upstream"
# Commits vs main
git log main..HEAD --oneline
# Changed files
git diff main..HEAD --statBlock if:
- On main/master
- No commits ahead of main
- Uncommitted changes (warn, offer to commit first)
2. Analyze Changes
Determine type from files:
feat— New functionalityfix— Bug fixesdocs— Only .md, docs/ changesrefactor— Restructure, same behaviortest— Test files onlybuild— build.gradle.kts, Makefile, CIchore— Maintenance
Determine scope from paths: use the scope table in the commit skill (compiler, codegen, gradle, annotations, samples, claude, ci).
3. Generate PR Title
<type>[(scope)]: <description>- Max 72 characters
- No trailing period
- Imperative mood
4. Populate PR Template
cat .github/pull_request_template.mdFill template with analysis from step 2.
5. Add the AI Attribution Footer
When Claude wrote or co-wrote the change, end the description with an attribution line after the template:
🤖 Generated with [Claude Code](https://claude.com/claude-code)If the session already supplies attribution lines for pull requests, use those verbatim instead. Keep the title free of attribution.
6. Push and Create Draft PR
# Push branch
git push -u origin $(git branch --show-current)
# Create draft PR (if `gh` isn't available, use the session's GitHub tool with the same draft, title and body)
gh pr create \
--draft \
--title "<title>" \
--body "$(cat <<'EOF'
<populated template>
EOF
)"7. Output
PR CREATED (Draft)
URL: {url}
Title: {title}
To mark ready: gh pr ready {number}
To view: gh pr view {number} --webError Messages
Not on feature branch:
PR BLOCKED — Cannot create PR from main/master
Create a feature branch first: git checkout -b feat/your-featureNo commits:
PR BLOCKED — No commits ahead of mainTitle too long:
PR BLOCKED — Title exceeds 72 characters (currently: {n})
Suggested: "{shortened}"Related Skills
commit— Clean commits before PRbdd-test-runner— Run tests before PR
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
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.

