AGENT SKILL

postmortem

A blameless incident postmortem: summary with impact numbers, a timeline, root cause and contributing factors, what went well and badly, and action items with owners.

License 许可
MIT
最近更新
2026-11-14
Stars
★ 1.1k
Forks
104

01 前言 Front Matter

od
{ "mode": "prototype", "craft": { "requires": [ "typography", "typography-hierarchy", "color", "accessibility-baseline" ] }, "preview": { "type": "html", "entry": "index.html" }, "platform": "desktop", "scenario": "engineering", "design_system": { "requires": false }, "example_prompt": "Write a postmortem for this incident from the notes and timeline I paste: impact, timeline, root cause, contributing factors, what went well and badly, and action items with owners." }
name
postmortem
tags
[ "postmortem", "incident", "retrospective", "sre", "outage", "事故复盘" ]
emoji
🩺
en_name
Postmortem
zh_name
事故复盘
category
doc
scenario
engineering
triggers
[ "postmortem", "post-mortem", "incident review", "incident report", "outage report", "事故复盘" ]
description
A blameless incident postmortem: summary with impact numbers, a timeline, root cause and contributing factors, what went well and badly, and action items with owners.
en_description
A blameless incident postmortem: summary with impact numbers, a timeline, root cause and contributing factors, what went well and badly, and action items with owners.

02 资源文件 1

03 SKILL.md 正文

Postmortem

Intent. A blameless account of an incident that helps the team learn and prevents a repeat. It describes systems and decisions, never blames people.

Ground it

Write only what the sources say. Gather them yourself before writing a word:

  • the incident notes, chat or pager timeline, and dashboards the user provides. Ask for them if they're missing: never invent an incident, a timestamp or an impact number;
  • the commits that caused and fixed it (git log --since … --until …, git show <sha>), and the code paths involved.

Keep the list of files, commits and refs you read: they go in a Sources section at the bottom of the page and in sources when you register (register_open_design_artifact with kind html). If something the document needs isn't in the sources, ask the user, or mark it plainly as an open question. Never invent numbers, dates, names or quotes.

Structure

  1. Header: title, incident id if any, date, a severity pill (SEV1–SEV4), status (data-od-status="draft|reviewed|closed"), authors.
  2. Summary: what happened, who was affected and for how long, in three sentences.
  3. Impact: a row of 2–4 big numbers (duration, users or requests affected, error budget used), each with its source.
  4. Timeline: a table of timestamps (with the timezone) and events: detection, escalation, mitigation, resolution.
  5. Root cause: the technical chain, step by step. Link the commits.
  6. Contributing factors: what made it worse or slower to fix.
  7. What went well and What went badly: two short lists.
  8. Action items: a table of action, owner, priority and due date, each tied to a cause above.
  9. Sources.

Use blameless language: "the deploy pipeline allowed…", not "X forgot…".

Design

  • One readable column (max-width 720–780px), body 16–17px with line-height 1.55–1.65, generous space between sections. It must read well on a phone: tables sit in an overflow-x: auto wrapper, nothing has a fixed width wider than the screen.
  • Style from the active design system (--bg, --surface, --fg, --muted, --border, --accent); status colours are the only extra colours (green for accepted/done, amber for proposed/in progress, red for rejected/high severity, grey for superseded).
  • Headings are short labels; the content does the talking. Code, paths and commands in a monospace face, with overflow-wrap: anywhere so long paths wrap on phones.
  • Finish with check_open_design_artifact and fix every error at desktop and mobile width.

04 同仓库更多 Skills

feimacode
feimacode/open-design-agent-kit

youtube-clipper

YouTube clip generation and editing with automated workflows — pull source video, slice highlights, add captions, and export.

★ 1.1k
feimacode
feimacode/open-design-agent-kit

writing-guidelines

Review docs/prose for Writing Guidelines compliance. Use when asked to "review my docs", "check writing style", "audit prose", "review docs voice and tone", or "check this page against the writing handbook".

★ 1.1k
feimacode
feimacode/open-design-agent-kit

wpds

WordPress Design System. Apply WordPress's official design tokens, typography, and component patterns to themes and sites.

★ 1.1k
feimacode
feimacode/open-design-agent-kit

weread-year-in-review-video-template

WeRead-inspired HyperFrames video template for vertical annual reading reports, personal reading dashboards, book-note recaps, and shareable year-in-review stories. Use when users want a 9:16 HTML-to-MP4 reading report with warm paper texture, editorial Chinese typography, book-page metaphors, data highlights, and deterministic motion.

★ 1.1k
feimacode
feimacode/open-design-agent-kit

web-design-guidelines

Review UI code for Web Interface Guidelines compliance by the Vercel engineering team. Covers layout, typography, color, motion, and accessibility for product UI.

★ 1.1k
feimacode
feimacode/open-design-agent-kit

web-clone

网站复刻 / 克隆方法论。USE WHEN 用户说 复刻网站、克隆网站、clone website、抄个站、仿站、 照着这个站做一个、reproduce site、还原某个网页效果、把这个站搬下来改成我的、 复刻某个交互/WebGL/Canvas/Three.js 效果。提供「先拿真源码 → 判路径 → 逆向拆解 → 搭工程 → 替换内容」的可移植决策树,覆盖静态站 / React-Vue-Next 内容站 / WebGL-Canvas 重前端站三大分支,并强制核对任何 AI 二手分析里的可执行代码。

★ 1.1k