optimizationpostmortem

edwardmonteiro's avatarfrom edwardmonteiro

Produce a post-incident analysis capturing timeline, root causes, impact, and corrective actions.

⭐0stars🔀1forks📁View on GitHub🕐Updated Dec 23, 2025
[Incident Response]

When & Why to Use This Skill

This Claude skill streamlines the post-incident analysis process by generating structured, professional postmortems. It helps engineering and SRE teams document incident timelines, conduct deep-dive root cause analyses (RCA), and establish clear corrective actions to enhance system reliability and organizational learning.

Use Cases

  • Post-outage documentation: Automatically generate a detailed report after a system failure to capture the sequence of events, impact scope, and resolution steps.
  • Root Cause Analysis (RCA): Utilize structured methodologies like the '5 Whys' or fishbone diagrams to identify underlying technical or process-related triggers for service degradations.
  • Reliability Improvement: Identify monitoring gaps and define preventive measures with assigned owners and deadlines to reduce the likelihood of recurring incidents.
  • Stakeholder Communication: Create standardized summaries of incidents for leadership and cross-functional teams to ensure transparency and accountability across the organization.
nameoptimization.postmortem
phaseoptimization
descriptionSteps taken during mitigation.
- namemitigation_steps

Purpose

Standardize postmortem creation so engineering and reliability teams can rapidly document learnings and accountability.

Pre-run Checklist

  • ✅ Gather incident alerts, logs, and chat transcripts.
  • ✅ Confirm timeline with responders.
  • ✅ Align on severity classification and stakeholders.

Invocation Guidance

codex run --skill optimization.postmortem \
  --vars "incident_summary={{incident_summary}}" \
         "impact_scope={{impact_scope}}" \
         "duration={{duration}}" \
         "mitigation_steps={{mitigation_steps}}"

Recommended Input Attachments

  • Incident timeline exports (PagerDuty, Opsgenie, Slack).
  • Monitoring dashboards or screenshots.
  • Customer communications sent during the incident.

Claude Workflow Outline

  1. Summarize incident context, severity, impact, and timeline.
  2. Detail sequence of events with timestamps and decision points.
  3. Perform root cause analysis using techniques like 5 Whys or fishbone.
  4. Document detection gaps and monitoring improvements.
  5. Provide corrective and preventive actions with owners and deadlines.

Output Template

# Postmortem — {{incident_summary}}

## Impact & Timeline
- Severity:
- Impact Scope:
- Duration:
| Time | Event | Owner | Notes |
| --- | --- | --- | --- |

## Root Cause Analysis
- Primary Cause:
- Contributing Factors:
- Detection Gaps:

## Corrective Actions
| Action | Owner | Due Date | Status |
| --- | --- | --- | --- |

## Preventive Measures
- Monitoring Updates:
- Process Improvements:
- Follow-up Reviews:

Follow-up Actions

  • Review postmortem in reliability or engineering forums.
  • Track actions in the incident management system.
  • Schedule follow-up to verify preventive measures are in place.