agentleFS
Sign inStart free
frankxaifrankxai/starlight-agent-skills/skills/coding/starlight-ship/SKILL.md

starlight-ship

Skill1 starsChanged 4 months agoMITView on GitHub

Prepare an authorized change for review or landing by checking repository state, required gates, secrets, diff scope, deployment evidence, rollback, and merge policy. Use when a branch or pull request is ready for a release decision; never rewrite, merge, deploy, or publish without the authority granted for that action.

1 other project uses a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

4 rules. About 0 are common, 4 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈4

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 0

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Verify path, repository root, origin, branch, and working-tree ownership.

    line 27
  2. Distinctive:

    Run proportionate tests, lint/type gates, secret scan, and diff checks.

    line 29
  3. Distinctive:

    Use a preview deployment for web changes when the repository supports it.

    line 30
  4. Distinctive:

    Use non-interactive Git, the repository’s CI, secret scanner, and hosting connector.

    line 41

≈ “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

Starlight Ship

Purpose

Turn “the code seems done” into a reproducible release decision while preserving unrelated work and repository governance.

When it fires

  • A change is ready to commit, push, open for review, merge, or deploy.
  • The user asks for release, landing, PR, or Git/Vercel operations.
  • A stale or failing PR needs evidence-based disposition.

Inputs

  • Repository instructions, branch policy, and required checks.
  • Exact diff, test results, review findings, deployment state, and rollback path.
  • User authority for commit, push, merge, production, publication, and cleanup.

Workflow

  1. Verify path, repository root, origin, branch, and working-tree ownership.
  2. Inspect the complete diff and exclude unrelated user changes.
  3. Run proportionate tests, lint/type gates, secret scan, and diff checks.
  4. Use a preview deployment for web changes when the repository supports it.
  5. Record independent review findings and resolve blockers.
  6. Confirm mergeability, required checks, release notes, rollback, and post-merge monitoring.
  7. Execute only the authorized Git or deployment action; never force-push or rewrite shared history by default.

Output contract

Return revision, diff scope, gates and results, review status, preview/production state, known limitations, rollback, action taken, and remaining human decisions.

Tools & MCP

Use non-interactive Git, the repository’s CI, secret scanner, and hosting connector. Preserve dirty worktrees, avoid destructive reset/checkout commands, and stop session-owned servers.

Quality bar

  • “Green” means required gates passed on the exact head revision.
  • Skipped or unavailable checks are explicit.
  • A successful preview is not called a production release.
  • Merge or deployment follows branch protection and human authority.

Example

Input: “Land PR 21 if it is ready.”

Good output: verify the exact head, required CI and security reviews, run release smoke tests against the preview, resolve comments, merge using repository policy, verify production, and retain a rollback revision.

Built on SIP — Starlight Intelligence Protocol Vertical: starlight-agent-skills · portable capability layer

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