ship
Publish finished work as a pull request and drive its CI checks to green.
Skill by taecontrol
Install
Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.
npx agentmarket install skill/taecontrol-ship
Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.
Get the weekly diff
The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.
About
Use cases: devops
Reviews
Write a reviewNo reviews yet. Installed it? Reviews come from verified installers only. How reviews work
Permissions
What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.
- Tools
- No tools found
- Secrets
- None asked for
- Install scripts
- None found
Network hosts it talks to 1
codeload.github.com· found in install command
Files it touches
~/.claude/skills/ship
What's inside
- Skill name
ship- Description
- Publish finished work as a pull request and drive its CI checks to green. Use when work is ready for review, or when asked to open a PR or make its checks pass.
- License
- MIT
- Instructions
- 32 lines in SKILL.md · view on GitHub
Files (2)
- SKILL.md
- openai.yaml · agents/
SKILL.md (first lines)
# Ship A pull request is ready when the human can merge it without asking anything. Green checks in CI, not on this machine, are the bar. ## Prepare The branch holds only durable work: temporary artifacts such as specs, prototypes, and reports stay out of it. When the work came from a spec, check that what must outlive it is in the change, not only in the spec. Bring the branch up to date with its base, and rerun the project's gates when that changed anything. ## Open Push the branch and open the pull request against its base, following the project's conventions for titles and descriptions, in the language its team reads. Write the description for the **reviewer**: someone who knows the codebase but not how this change was made, and who wants to understand it and review it quickly. - **What and why:** the problem in a few sentences, the issue it closes, and the plan when there is one. - **How to review:** where to start, the order to read the change in, where the risk sits, and what can be skimmed, such as generated code, mechanical moves, or fixtures. - **How to try it:** the steps to see it work, with screenshots or recordings where a person must look, attached where the platform allows. - **Decisions worth knowing:** choices made while building it that a reviewer would otherwise question, in the product's terms. - **Not included:** what is real but was left out, with its follow-up. Leave out empty sections. Describe the change, not how it was made, and cite only what the reviewer can open: the spec's example IDs, review rounds, deviations from the spec, and working files belong in the delivery's report to the human. Run `unslop` on the description before publishing it. ## Drive checks to green Watch every CI check until it finishes. For each failure, find why it fails in CI when it passed locally, such as a different platform, browser, path, timeout, or concurrency, fix the cause, and push again. A check that passes only on a retry is flaky, not green: diagnose it with `diagnosing-bugs`. Make checks pass by fixing code, tests, or configuration; disabling a check or weakening a test to get green does not count. When a failure is outside the change, show that it fails on the base too, say so in the pull request, and find or open its issue so the next delivery inherits the diagnosis. When a failure needs something you cannot get, such as a secret, infrastructure access, or a human decision, stop and name it. ## Done Every CI check is green, the description matches the final state, and the pull request is ready to merge. The human merges.
From taecontrol/skills, MIT.
More from taecontrol/skills
- 0.592score
strategic-programming
taecontrol
Apply strategic programming to non-trivial implementation and repair work when green tests alone cannot establish a maintainable design.
SkillAFreshness: A★ 1Claude Code +9 - 0.592score
architect
taecontrol
Settle a one-way door before code locks it: a change to persisted data or its schema, a contract that writes, a security boundary, an ownership model, or a…
SkillAFreshness: A★ 1Claude Code +9 - 0.592score
prototype
taecontrol
Make a design question tangible before committing to it: compare production-representative UI directions for a screen, component, or short flow, or drive a…
SkillAFreshness: A★ 1Claude Code +9 - 0.589score
refine
taecontrol
Review a working change with fresh eyes against its spec, then make it right and make it fast without changing the behavior its acceptance tests lock, in at…
SkillAFreshness: A★ 1Claude Code +9 - 0.587score
shape
taecontrol
Shape a problem with the human into one accepted spec: what to build, what success means, and how to prove it, before anyone writes production code.
SkillAFreshness: A★ 1Claude Code +9 - 0.587score
research
taecontrol
Investigate a bounded question against high-trust primary sources through parallel independent research, then produce a cross-checked Markdown report.
SkillAFreshness: A★ 1Claude Code +9 - 0.587score
spike
taecontrol
Run a bounded technical experiment when observation is needed to settle feasibility, integration, performance, or tool behavior.
SkillAFreshness: A★ 1Claude Code +9 - 0.587score
setup-verification
taecontrol
Give agents hands and eyes on a project: one way to launch, drive, and observe the real product, and acceptance tests that run fast and the same locally and…
SkillAFreshness: A★ 1Claude Code +9 - 0.587score
deliver
taecontrol
Deliver an accepted spec autonomously, from failing acceptance tests to a pull request with green checks.
SkillAFreshness: A★ 1Claude Code +9 - 0.587score
diagnosing-bugs
taecontrol
Establish the mechanism behind a hard, intermittent, recurring, or performance failure from a faithful reproduction or distinguishing runtime evidence.
SkillAFreshness: A★ 1Claude Code +9 - 0.587score
garden
taecontrol
Tend a codebase between features: find drift in one area, group it by cause, and fix the causes worth their cost in one behavior-preserving pull request.
SkillAFreshness: A★ 1Claude Code +9 - 0.584score
skill-guide
taecontrol
Explain available skills, compare their boundaries, or recommend what the user should invoke next.
SkillAFreshness: A★ 1Claude Code +9
Signals
- Stars
- 1
- Forks
- 0
- Open issues
- 0
- Last commit (this folder)
- Oct 9, 2026 (20 days ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- MIT
- Repository
- github.com/taecontrol/skills
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
More from taecontrol
- 0.592score
strategic-programming
taecontrol
Apply strategic programming to non-trivial implementation and repair work when green tests alone cannot establish a maintainable design.
SkillAFreshness: A★ 1Claude Code +9 - 0.592score
prototype
taecontrol
Make a design question tangible before committing to it: compare production-representative UI directions for a screen, component, or short flow, or drive a…
SkillAFreshness: A★ 1Claude Code +9 - 0.592score
architect
taecontrol
Settle a one-way door before code locks it: a change to persisted data or its schema, a contract that writes, a security boundary, an ownership model, or a…
SkillAFreshness: A★ 1Claude Code +9 - 0.589score
refine
taecontrol
Review a working change with fresh eyes against its spec, then make it right and make it fast without changing the behavior its acceptance tests lock, in at…
SkillAFreshness: A★ 1Claude Code +9 - 0.587score
deliver
taecontrol
Deliver an accepted spec autonomously, from failing acceptance tests to a pull request with green checks.
SkillAFreshness: A★ 1Claude Code +9 - 0.587score
garden
taecontrol
Tend a codebase between features: find drift in one area, group it by cause, and fix the causes worth their cost in one behavior-preserving pull request.
SkillAFreshness: A★ 1Claude Code +9
Cite this page
ship skill. Agent Mart. https://agentmart.dev/skills/taecontrol-ship (last updated 2026-10-24)