Skip to content
Back to skills

Ship

ASecurity

End-to-end delivery of an authorized, clearly scoped task: runs the three skills $implement, $review, $cleanup in sequence. Not used when scope or acceptance is unclear or the root cause is unknown; adds no new gates or repeated confirmations.

  • 3 stars
  • 0 votes
  • 0 copies
  • 2 views
  • Added September 24, 2026
ai-agentsgo

Security analysis

A100/100

Scanned September 24, 2026

npx -y skills add YSAA1/harness-workflow --skill ship --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Ship?

Add the live security badge to your README. It updates with every re-scan.

Security grade badge for Ship
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/ysaa1-ship/badge)](https://www.skillsdirectory.com/skills/ysaa1-ship)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
SKILL.md
---
name: ship
description: "End-to-end delivery of an authorized, clearly scoped task: runs the three skills $implement, $review, $cleanup in sequence. Not used when scope or acceptance is unclear or the root cause is unknown; adds no new gates or repeated confirmations."
---

# One delivery: $implement -> $review -> $cleanup

Carry an authorized task through modification, review, and close-out within one session. This skill only orchestrates ordering and abort conditions; each stage's protocol defers to the corresponding skill.

## Process

1. Confirm preconditions: scope and acceptance are clear, and the task is authorized to run the full course. Unclear scope routes to $brainstorm/$plan; unknown failure root cause routes to $diagnose, returning here to continue once resolved.
2. Run $implement: minimal changes + test-driven on agreed seams + targeted checks. Low-risk mechanical changes may proceed to close-out after self-review per its completion contract.
3. Substantive-risk changes (cross-module, permission/data boundaries, deployment, explicit user request) continue to $review: structural review + fresh evidence adjudication; Critical/Important findings go back to modification, with the affected parts re-verified.
4. Run $cleanup: tidy documents, code, and leftovers so the recovery surface forms a state the next session can continue from. When the task closes in this round, the retirement contract is included: within the same commit that flips this track's row out of active, complete flip the row → lessons inheritance → delete this track's documents (including unapproved/abandoned ones) → state sync.
5. Output a delivery summary: change overview, evidence and conclusions, close-out result, remaining items.

## Sequencing discipline

- A later stage does not overturn a previous one without reason; when review finds something to change, redo only the affected parts.
- Do not skip cleanup to finish early, nor expand cleanup into a repo-wide tidy-up.
- Any stage hitting an authorization boundary or a trade-off that needs the user's decision: stop and explain; do not decide for the user.

## Recommended next skill

- Delivery complete: end.
- Unresolved findings or blocked: record and hand off along the corresponding lane.

Attribution

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments

Loading comments…

Newsletter

Get the weekly list of safe new skills, and the ones we pulled.