agentleFS
Sign inStart free
jaydubya818jaydubya818/MissionControl/skills/factory-ship/SKILL.md

factory-ship

Skill2 starsChanged 8 months agoMITView on GitHub

Experimental workflow for publishing and completing configured software delivery work. Use when the user or an enabled factory policy asks to ship.

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
  1. Very specific:

    Read .agent-factory/config.yaml, the optional skill_prompts.factory-ship entry, and the repository's instructions before publishing.

    line 15
  2. Distinctive:

    Use a clean automation-owned worktree when the scheduler provides one.

    line 23
  3. Distinctive:

    Verify changes: Run the configured formatter, tests, and release checks.

    line 25
  4. Distinctive:

    Verify deployment only when configured.

    line 37

≈ “Common” is an estimate: at least 1 in 5 of the 308,232 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

Factory Ship

Start with the Factory guide for workflow setup and configuration.

Read .agent-factory/config.yaml, the optional skill_prompts.factory-ship entry, and the repository's instructions before publishing. See the Factory configuration reference. The prompt adds project guidance; editing code does not by itself authorize a PR update, approval, merge, or deployment.

Delivery steps

  1. Confirm ownership. Identify the target repository, task-owned worktree, branch, and complete task-related change set. Preserve unrelated changes. Use a clean automation-owned worktree when the scheduler provides one; never take over a peer's checkout.
  2. Verify changes. Run the configured formatter, tests, and release checks. Report checks that were skipped or unavailable.
  3. Publish when enabled. Open or update a PR only when authorized. Apply configured title, body, draft status, labels, and communication rules. Do not tag or message people unless enabled.
  4. Resolve feedback. Compare comments with the code and current human direction. Make one coherent update, then rerun affected checks.
  5. Apply merge gates. Merge only under its own enabled policy and while all criteria hold on the unchanged live head. Use a head-match guard when the host supports it. A head change invalidates tied evidence and resets the soak.
  6. Verify and close out. Confirm the merge in the current base branch. Verify deployment only when configured; a merge is not live proof. Close issues, rotate worktrees, and notify only at their configured proof points.

Stop conditions

Hold the work when ownership, authorization, a required check, or live host state is missing or unclear. State the exact next step; do not turn an inconclusive result into success.

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.

Reviews can't be read right now.

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