postmortem-writing
Write Blameless Postmortems
๋ํ ๋ค์์์ ์ฌ์ฉํ ์ ์์ต๋๋ค: wshobson
Turn incidents into organizational learning without blame. Create structured postmortems that identify root causes and prevent recurrence.
์คํฌ ZIP ๋ค์ด๋ก๋
Claude์์ ์ ๋ก๋
์ค์ โ ๊ธฐ๋ฅ โ ์คํฌ โ ์คํฌ ์ ๋ก๋๋ก ์ด๋
ํ ๊ธ์ ์ผ๊ณ ์ฌ์ฉ ์์
ํ ์คํธํด ๋ณด๊ธฐ
"postmortem-writing" ์ฌ์ฉ ์ค์ ๋๋ค. Help me write a postmortem. Payment service was down for 47 minutes after deployment v2.3.4. Root cause was database connection pool exhaustion.
์์ ๊ฒฐ๊ณผ:
Generates a complete postmortem with: Executive Summary with impact metrics, detailed Timeline table with timestamps, Root Cause Analysis using 5 Whys, Detection section covering what worked and gaps, Response evaluation, Impact assessment (customers, business, technical), Lessons Learned, and Action Items table with priorities and owners
"postmortem-writing" ์ฌ์ฉ ์ค์ ๋๋ค. Conduct 5 Whys for: API latency spiked after cache flush
์์ ๊ฒฐ๊ณผ:
Provides structured 5 Whys analysis starting with problem statement, then drilling through each why with evidence, identifying root causes (missing cache warming, no partial invalidation), and systemic improvements with prevention and detection strategies
๋ณด์ ๊ฐ์ฌ
์์ Static analysis flagged 34 patterns that are all false positives. The backtick patterns are Markdown code block delimiters, not shell command execution. The URLs are reference links to external documentation, not network calls. This skill contains only documentation and templates for writing postmortems.
ํ์ง ์ ์
๋ง๋ค ์ ์๋ ๊ฒ
Post-Incident Review
After resolving a production incident, use this skill to structure the postmortem document with timeline, impact assessment, and root cause analysis.
Incident Response Training
Use the templates and guides to train new team members on effective incident documentation and blameless review practices.
Organizational Learning
Create standardized postmortems that can be shared across teams to prevent similar incidents and spread learnings.
์ด ํ๋กฌํํธ๋ฅผ ์ฌ์ฉํด ๋ณด์ธ์
Help me write a postmortem for an incident that occurred on [DATE]. The service [SERVICE NAME] was down for [DURATION] affecting [NUMBER] users. The issue was caused by [BRIEF DESCRIPTION]. Generate a postmortem document with executive summary, timeline, and initial root cause analysis.
Conduct a 5 Whys analysis for this incident: [INCIDENT DESCRIPTION]. Start with the problem statement and drill down through at least 5 levels of why questions to identify root causes.
I'm facilitating a postmortem meeting for [INCIDENT SUMMARY]. Create a meeting agenda with time boxes, key discussion questions, and tips for maintaining a blameless environment during the review.
Based on this incident analysis [PASTE ANALYSIS], generate prioritized action items. For each item include: priority level, specific action, recommended owner role, estimated timeline, and whether it prevents, detects, or mitigates future occurrences.
๋ชจ๋ฒ ์ฌ๋ก
- Start the postmortem within 24 hours while memories are fresh
- Focus on what conditions allowed the failure, not who made mistakes
- Assign every action item to a specific owner with a due date
ํผํ๊ธฐ
- Naming individuals as the cause of the incident
- Creating action items without owners or deadlines
- Skipping postmortems for minor incidents that reveal patterns
์์ฃผ ๋ฌป๋ ์ง๋ฌธ
What incidents require a postmortem?
How do I maintain a blameless culture?
When should the postmortem meeting be held?
Who should attend the postmortem meeting?
What makes a good action item?
Should we postmortem near-misses?
๊ฐ๋ฐ์ ์ธ๋ถ ์ ๋ณด
์์ฑ์
sickn33๋ผ์ด์ ์ค
MIT
๋ฆฌํฌ์งํ ๋ฆฌ
https://github.com/sickn33/antigravity-awesome-skills/tree/main/skills/postmortem-writing์ฐธ์กฐ
main
ํ์ผ ๊ตฌ์กฐ
๐ SKILL.md