agentleFS
HenryQWHenryQW/skills/pr-launchpad/SKILL.md

pr-launchpad

Skill1 starsChanged 4 months agoApache-2.0View on GitHub

Create or publish a GitHub or GitLab pull request when asked to open or create a PR. Use when must inspect the actual diff, commit pending work with Conventional Commits, push the current branch, and create a PR with a Conventional Commit title and fixed body template.

  • Commits and pushes

The file

We split this skill into the rules it gives an agent

3 rules. About 0 are common, 1 is distinctive, and 2 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈1

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 2

    Very specific

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

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

    Write the multiline body to a temporary file or heredoc.

    line 40
  2. Very specific:

    After create or reuse, when shipyard_manifest is present, require:

    line 47
  3. Distinctive:

    Use one closure bullet per clearly resolved issue.

    line 72

≈ “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

pr-launchpad

Publish the current branch as a pull request.

Inputs and Memory

  • base_branch defaults to the repository default branch.
  • issue_number or issue_numbers may come from input, branch, or title.
  • validated_head plus non-empty validation commands/results may come from nested issue-workbench.
  • shipyard_manifest supplies only close targets, child heads, checks/skips, and review status.

Direct invocation owns $agent-memory; nested invocation defers it to issue-workbench or shipyard.

Procedure

  1. Resolve the base, preferring its remote-tracking ref. Inspect git status --short, git log --oneline <base>...HEAD, the committed branch diff, and staged and unstaged diffs. Do not commit .context/, .agents/, or local progress. Stop when unrelated changes make intended scope ambiguous; otherwise make the focused Conventional Commit.
  2. Reuse HEAD-bound validation before running a check. With a manifest, first run python3 <shipyard_dir>/scripts/manifest.py --manifest <shipyard_manifest> can-reuse $(git rev-parse HEAD) and reuse recorded evidence only on success. Without a manifest, reuse nested Workbench evidence only when validated_head equals current HEAD and validation is non-empty. Otherwise run the smallest relevant non-destructive validation; state honestly when no relevant check exists.
  3. Derive a Conventional Commit PR title and the fixed body below from the live diff and validation. Manifest evidence never overrides git.
  4. Resolve the upstream and its SHA. Push with git push --set-upstream origin HEAD when absent, git push when its SHA differs from HEAD, and do not push when they already match.
  5. Write the multiline body to a temporary file or heredoc. Before creating, query open PRs/MRs whose source is the current branch (gh pr list --head <branch> or the glab equivalent). Reuse exactly one only when its source branch, resolved base, and issue/diff scope match; refresh its title/body to the fixed values when needed. If multiple candidates exist, or a same-branch request has a different base or scope, stop instead of creating a duplicate. Otherwise create against the resolved base.
  6. After create or reuse, when shipyard_manifest is present, require:

    python3 <shipyard_dir>/scripts/manifest.py --manifest <shipyard_manifest> set-pr <url>

    An explicitly waived or replaced unavailable external review check is recorded as Pending external unavailable check: <check> in caller/progress context, not routed to repair skills.

  7. Compact .context/progress.md without suppressing a created or reused PR URL.
## Summary

- <2 to 4 bullets covering the main changes>

## Testing

- <commands actually run or an honest reason none applied>

## Close Issue

- Closes #<issue_number>

Use one closure bullet per clearly resolved issue; otherwise omit the section. Reply with only the PR URL.

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.