Incident Postmortem Repoter

The problem

Engineering teams often struggle to produce objective, blameless postmortems after an outage. Manually reconciling messy Slack logs, monitoring alerts, and conflicting timestamps leads to vague reports that fail to prevent future incidents.

What it does

• Synthesizes raw incident logs, alert messages, and team observations into a structured postmortem.
• Constructs a chronological event timeline that flags conflicting timestamps and identifies the first abnormal signal.
• Performs a multi-layered Root Cause Analysis identifying triggers, contributing factors, and systemic gaps.
• Generates a prioritized table of actionable follow-up tasks with specific verification methods.
• Drafts internal and customer-facing incident communications based on verified facts only.

Why this beats prompting it yourself

This skill enforces a strict blameless culture, focusing on system failures rather than human error. It prevents the common AI habit of hallucinating metrics or timestamps by explicitly labeling unknowns and requiring evidence-based causality.

Use cases

• Transforming a chaotic "war room" Slack export into a professional engineering report.
• Conducting a 5 Whys analysis to move beyond the immediate trigger to the systemic cause.
• Creating customer-facing status page updates that balance transparency with security.
• Auditing monitoring gaps to determine why automated alerts failed to trigger.

Known limitations

It cannot verify the truth of provided logs or invent missing data. Qualitative impact assessment requires user-provided metrics.