Skip to main content

document-a-known-issue

Create a maintainable known-issue record with verified symptoms, affected scope, safe workarounds, status, and update ownership. Use when support, customers, or operators need a reliable reference for an unresolved defect.

来源信息

仓库
protosphinx/sphinxstack
最近来源活动
2026年7月29日 13:55
检测到的 SKILL.md 语言
英语
星标
3
Forks
0

安装方式

默认使用会先检查来源的 Prompt;你也可以切换为直接命令,或下载本地副本。

检查来源文件

决定是否安装前,请先阅读 SKILL.md,以及 SkillsMP 当前展示的配套文件。

文件资源管理器
2 个文件

正在显示 SKILL.md

SKILL.md
来源说明 · 只读预览
name
document-a-known-issue
category
write
description
Create a maintainable known-issue record with verified symptoms, affected scope, safe workarounds, status, and update ownership. Use when support, customers, or operators need a reliable reference for an unresolved defect.
# document-a-known-issue Publish the smallest accurate record that helps people recognize the problem and respond safely. Keep confirmed behavior separate from suspected cause and update the page as evidence changes. ## Inputs - Gather the defect or incident record, reproduced symptoms, affected versions, start time, and current status. - Confirm any workaround with engineering, security, support, and product owners as relevant. - Decide which details belong in public, customer-specific, internal, and security-restricted versions. ## Procedure 1. Write a searchable title using the visible symptom rather than an unproven root cause. 2. Describe expected behavior, observed behavior, and the conditions required to reproduce it. 3. State affected and unaffected products, versions, platforms, regions, or account types with evidence. 4. Add a safe workaround with prerequisites, trade-offs, reversal steps, and verified result. 5. Record severity, first observed time, current status, owner, and next update time. 6. Link the canonical defect or incident record and related support guidance. 7. List signals that indicate a different problem and where those cases should go. 8. Add a dated update log and explicit resolved, superseded, or retired criteria. ## Boundaries Never publish credentials, exploit details, customer data, private incident discussion, or a suspected person at fault. Do not claim a root cause, fix, or affected scope before it is verified. Avoid unsafe workarounds. ## Done - The issue can be found by its verified symptom and distinguished from similar failures - Scope, status, workaround, owner, and next update are current and checked - Public and restricted details are separated according to privacy and security rules - The record links to evidence and includes a tested retirement or resolution condition
在 GitHub 查看