incident-postmortem

Incident Postmortem Skill

This skill produces a complete, blameless incident postmortem document following industry-standard format. Output is ready to share with engineering teams, leadership, and affected stakeholders.

Required Inputs

Ask the user for these if not provided:

• Incident title / ID
• Severity (P1 / P2 / P3 or SEV1 / SEV2 / SEV3)
• Date and duration of the incident
• What happened (rough notes are fine — the skill will structure them)
• Services or systems affected
• Customer impact (how many users, what was degraded)
• How it was detected
• How it was resolved
• Initial thoughts on root cause
• Action items already identified (optional)

Output Structure

---

Incident Postmortem: [Incident Title]

Incident ID: [ID] Severity: [P1/P2/P3] Date: [Date] Duration: [Start time → Resolution time — total duration] Status: [Resolved / Monitoring / Ongoing] Author: [Leave blank for user to fill] Last updated: [Date]

---

Executive Summary

[3–5 sentences. Describe what happened, who was affected, and what was done to resolve it. Written for a non-technical stakeholder. No jargon. No blame.]

---

Impact

Dimension	Details
Users affected	[Number or percentage]
Services degraded	[List affected services]
Business impact	[Revenue, SLA breach, support tickets, etc. if known]
Duration	[Total time from first detection to full resolution]

---

Timeline

List events in chronological order. Each entry: [HH:MM UTC] — [What happened. Who did what. What changed.]

Rules for timeline entries:

• Use passive or system-focused language — avoid "X made a mistake"
• Include: first symptom, detection, escalation, hypothesis tested, fix applied, confirmation of resolution
• Note time between key events (e.g. "22 minutes between detection and escalation")

---

Root Cause

Primary root cause: [One clear sentence. Technical but plain. "A misconfigured deployment config caused..."]

Contributing factors:

• [Factor 1 — e.g. lack of canary deployment meant change hit 100% of traffic immediately]
• [Factor 2 — e.g. alert threshold was set too high to catch the initial degradation]
• [Factor 3 — add as many as are relevant]

Why did our existing safeguards not prevent this? [Honest paragraph explaining why monitoring, tests, or processes didn't catch this earlier. This is where blameless analysis matters most — focus on system gaps, not individual failures.]

---

Detection

• How was it first detected? [Customer report / automated alert / internal monitoring / manual observation]
• Time from incident start to detection: [X minutes]
• Should we have detected this faster? [Yes / No — and why]

---

Resolution

What fixed it? [Clear description of the actual fix — one paragraph] Why did this work? [Brief technical explanation] Was there a temporary mitigation before full resolution? [Yes/No — describe if yes]

---

Action Items

#	Action	Owner	Due Date	Priority
1	[Specific, testable action]	[Team or person]	[Date]	P1/P2/P3

Rules for action items:

• Each action must be specific enough to close as "done" or "not done" — no vague items like "improve monitoring"
• Distinguish between: Prevent recurrence (fix the root cause), Improve detection (catch it faster next time), Improve response (resolve it faster next time)
• Assign a real owner — not "team" or "TBD" if avoidable
• Flag P1 actions as items that block the incident from being marked fully closed

---

What Went Well

[3–5 honest observations about the response. Include: fast collaboration, good runbooks used, effective escalation, clear communication. This section builds team confidence and reinforces good habits.]

---

Lessons Learned

[3–5 key insights from this incident that are worth sharing beyond this team. Write these as transferable lessons — e.g. "Our runbook for database failover didn't account for read-replica lag. All runbooks involving database failover should be reviewed."]

---

Communication Log

[Optional — list external communications sent: status page updates, customer emails, support responses. Include timestamps.]

---

Quality Checks

• Timeline has no blame-focused language
• Root cause is specific (not "human error")
• Contributing factors explain the systemic gaps
• Every action item has an owner and due date
• "What went well" section is genuine, not token
• Executive summary is readable by non-technical leadership

Example Trigger Phrases

• "Write a postmortem for the [incident name] outage"
• "Help me write a P1 incident report"
• "Generate an RCA document for [service] going down on [date]"
• "Draft a blameless postmortem from these notes: [paste notes]"