AGENT SKILL

ship

Decide whether a change is ready to hand to a human, and open the pull request when it is. Use once the work is done and verified, before opening a PR or telling the developer the change is ready.

License 许可
BSD-3-Clause
所属仓库
urban233/Canon
最近更新
2026-10-01
Stars
★ 1
Forks
0

01 前言 Front Matter

name
ship
description
Decide whether a change is ready to hand to a human, and open the pull request when it is. Use once the work is done and verified, before opening a PR or telling the developer the change is ready.

03 SKILL.md 正文

Ship

  1. Call canon_ship. Its three checks -- plan satisfied, evidence green at this HEAD, independent reviewer verdict present -- are the only things that make a change ready.
  2. If ready is false, stop. Work through missing in order rather than opening a pull request that isn't actually ready.
  3. Before opening the pull request, consider whether this branch made a decision worth recording: use the decide skill's test (a real alternative seriously considered, and a consequence that outlives this branch). If it passes, run decide first, so the record lands in the same commit as the work. This is never a gate -- it's imposed on you, not on the developer, and never blocks shipping.
  4. If ready is true, open the pull request (gh pr create). The body reproduces verbatim the plan's ## Non-goals and the section that states its approach -- ## Approach, or whatever this plan calls it (## Design, ## Shape); plan mode names it differently from plan to plan, so never invent one that isn't there. It states the exact command canon_evidence reports as green and, when ship_evidence is present, its command and that it passed for this tree -- not a summary of the diff, which the diff already is. Title: short, present tense, matching the plan's done: header field when one is set.
  5. Never merge or close the pull request, and never approve it -- a human does that. This skill's job ends at "opened, ready, and here is the evidence."

04 同仓库更多 Skills

urban233
urban233/Canon

testing-craft

Provides Google-derived testing strategy, test-writing craft, and test-suite health reference, distilled from Software Engineering at Google (the Flamingo book), the Google Testing Blog, and Google's published testing-infrastructure research. Canon's `reviewer` subagent reads the writing-craft and strategy references as review criteria. Invoke directly to design a test strategy, audit an existing suite's health, or triage a flaky or brittle test.

★ 1
urban233
urban233/Canon

ship

Decide whether a change is ready to hand to a human, and open the pull request when it is. Use once the work is done and verified, before opening a PR or telling the developer the change is ready.

★ 1
urban233
urban233/Canon

review

Dispatch Canon's independent reviewer subagents for this branch and read their verdicts back from canon_review. Use once a change is built and verified, before shipping. This skill only dispatches and reads back -- the review itself runs inside the subagent, in a fresh context, and never in this session.

★ 1
urban233
urban233/Canon

review-change

暂无简介

★ 1
urban233
urban233/Canon

plan

Shape what plan mode drafts, so the plan Canon saves needs no follow-up. Use whenever drafting a plan for approval in plan mode, before writing it to .canon/plans/.

★ 1
urban233
urban233/Canon

frame

Surface the open questions before planning a multi-branch feature, and shape the feature plan mode drafts. Use before entering plan mode for work that obviously won't fit in one branch and one pull request -- never for a change that already does.

★ 1