AGENT SKILL

pr-body

The shape of a pull request description: the smallest visual that makes the change clear, evidence that it works, and a merge-danger call (one-way or two-way door, plus blast radius). Use when writing or rewriting a PR description, and when opening or updating a pull request.

License 许可
Apache-2.0
最近更新
2026-10-19
Stars
★ 0
Forks
1

01 前言 Front Matter

name
pr-body
description
The shape of a pull request description: the smallest visual that makes the change clear, evidence that it works, and a merge-danger call (one-way or two-way door, plus blast radius). Use when writing or rewriting a PR description, and when opening or updating a pull request.

02 资源文件 1

03 SKILL.md 正文

PR Body

Write the body with this template, skipping preambles and keeping prose brief:

<one or two sentences: what changes and why>

## Summary

<the smallest visual that makes the change clear>

## Evidence

- **Before:** <failing test, error output, screenshot>
- **After:** <passing test, correct output, screenshot>
- **Not verified:** <anything you couldn't check, or "nothing">

## Merge danger

**Door:** <one-way or two-way>
<optional: why>

**Blast radius:** <who or what a bad merge would affect>

Closes #<issue>   (Refs #<issue> for partial work; omit when there is none)

Scale it to the change: a typo fix or a docs tweak needs the lead sentence and the link, not a visual or a merge-danger call.

Summary: pick the smallest view

Choose whichever makes the key point clearest; one is usually enough:

  • Pseudocode for logic or an algorithm.
  • A call tree for runtime control flow.
  • A component tree for UI structure, with the state and module boundaries that matter.
  • A shallow file tree for file responsibilities or a broad refactor.
  • Mermaid for interaction, control flow, or data flow between parts.
  • A diff block when the point is what changed in a shape that already exists, matched to the topic: a component tree, a file layout, a call tree, or pseudocode with + and - lines.
  • A small table when the change is a set of parallel items (skills, endpoints, flags).
 on(save)
-  write content
+  if content is unchanged
+    return cached result
+  write new content
+  invalidate cache

Put each visual next to the sentence it supports, and keep only the calls, files, and boundaries needed to see the change.

Evidence

Show it working, before and after. A screenshot is the strongest evidence for a visual change; gh pr create can't attach images, so upload it first and link the hosted URL. Next best is execution: the exact test that failed and now passes, or command output before and after. Say what you didn't verify.

Merge danger

  • Door: a two-way door is cheap to walk back (revert and redeploy). A one-way door isn't: data migrations, destructive actions, published APIs, sent messages.
  • Blast radius: what a bad merge would affect, considered widely: consumers of an API, other teams' pipelines, layout on mobile, performance, installed users.

Done when the body has the sections the change warrants, the evidence names what was and wasn't verified, and the issue it implements is linked if there is one.

04 同仓库更多 Skills

dougborg
dougborg/harness-kit

teach

Teach yourself a topic over several sessions, with lessons, references, and a record of what you've learned.

教师与教育
★ 0
dougborg
dougborg/harness-kit

svg-logo-designer

Generate SVG logos with multiple concepts, layouts, and color variations.

图像与多媒体
★ 0
dougborg
dougborg/harness-kit

documentation-writer

Writes and revises human-facing documentation — READMEs, contributor and setup guides, reference pages, architecture notes, CLAUDE.md, and changelog entries. Covers picking the right document type for the question being asked, structuring for skimming, writing examples that actually run, and keeping docs from going stale. Use when creating or rewriting docs, when a README has drifted from the code, or when deciding where a piece of information belongs. For SKILL.md and agent files, use /skill-writer instead.

内容创作者写作与文案
★ 0
dougborg
dougborg/harness-kit

adhd-mode

Shapes every response for a reader with ADHD: the next action first, numbered steps, progress restated each turn, tangents held back, concrete time estimates, wins made visible. Stays on for the rest of the session until the reader says "stop adhd mode". On Claude Code, the harness-kit:adhd output style does the same without invoking this skill.

★ 0
dougborg
dougborg/harness-kit

wait-what

That last message didn't land: re-explain it with the missing context.

★ 0
dougborg
dougborg/harness-kit

to-questionnaire

Turn a decision you can't make alone into a questionnaire for the one person who can answer it, filled in async or together in a meeting.

★ 0