agentleFS
Sign inStart free
ProjectWajaProjectWaja/claude-skillcraft/skills/ship-cycle/SKILL.md

ship-cycle

Skill0 starsChanged 6 months agoMITView on GitHub

Enforces the full dev cycle — branch, plan, implement, test, review, commit — with no skipped steps.

The file

Show its rules and text

We split this skill into the rules it gives an agent

17 rules. About 0 are common, 15 are distinctive, and 2 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈15

    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
  • could be automatic: a setting could enforce it
  1. Distinctive:

    Every feature, fix, or refactor follows the complete shipping cycle.

    line 14
  2. Distinctive:

    Create a feature branch from the latest main/develop

    line 20
  3. Very specific:

    Use conventional naming: feat/, fix/, refactor/, docs/

    line 21
  4. Distinctive:

    Never work directly on main

    line 22
  5. Very specific:

    Follow the plan-first protocol (see plan-first skill)

    line 26
  6. Distinctive:

    If plan-first is not installed, produce at minimum: requirements, approach, acceptance criteria

    line 27
  7. Distinctive:

    Write code in small, reviewable increments

    line 31
  8. Distinctive:

    Follow existing project conventions (indentation, naming, file structure)

    line 33
  9. Distinctive:

    Do not introduce new patterns without flagging them

    line 34
  10. Distinctive:

    Write or update tests for every behavioral change

    line 38
  11. Distinctive:

    Run the existing test suite before considering implementation complete

    line 39
  12. Distinctive:

    If no test framework exists, at minimum verify the change manually and document how

    line 40
  13. Distinctive:

    Check for: security issues, unnecessary complexity, missing error handling at boundaries, hardcoded values

    line 45
  14. Distinctive:

    If a step fails, do not skip to the next step.

    line 56
  15. Distinctive:

    If tests fail, do not commit.

    line 57
  16. Distinctive:

    If you realize mid-implementation that the plan was wrong, go back to step 2.

    line 58
  17. Distinctive:

    Document any deviations from the plan in the commit message.

    line 59

≈ “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

Ship Cycle

Every feature, fix, or refactor follows the complete shipping cycle. No shortcuts.

The Cycle

1. Branch

  • Create a feature branch from the latest main/develop
  • Use conventional naming: feat/, fix/, refactor/, docs/
  • Never work directly on main

2. Plan

  • Follow the plan-first protocol (see plan-first skill)
  • If plan-first is not installed, produce at minimum: requirements, approach, acceptance criteria

3. Implement

  • Write code in small, reviewable increments
  • Each increment should compile/run without errors
  • Follow existing project conventions (indentation, naming, file structure)
  • Do not introduce new patterns without flagging them

4. Test

  • Write or update tests for every behavioral change
  • Run the existing test suite before considering implementation complete
  • If no test framework exists, at minimum verify the change manually and document how

5. Review

  • Self-review the diff before committing
  • Check for: security issues, unnecessary complexity, missing error handling at boundaries, hardcoded values
  • Flag anything you're uncertain about to the user

6. Commit & PR

  • Atomic commits with conventional commit messages
  • PR description includes: what changed, why, how to test, screenshots if UI
  • Link to related issues if they exist

Rules

  • If a step fails, do not skip to the next step. Fix the failure first.
  • If tests fail, do not commit. Fix the tests or the code.
  • If you realize mid-implementation that the plan was wrong, go back to step 2.
  • Document any deviations from the plan in the commit message.

Escape Hatches

  • Hotfix mode: User says "hotfix" — compress to: implement, test, commit. Still no skipping test.
  • Prototype mode: User says "prototype" or "spike" — compress to: branch, implement, commit. Mark the branch as experimental.

These modes must be explicitly requested. Default is always the full cycle.

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