Production Incident Triage Starter

The problem

Initial incident reports are often a chaotic mix of conflicting logs, team theories, and fragmented timestamps. Without a clear triage, engineers waste time chasing red herrings or confusing correlation with actual causality.

What it does

• Separates observed facts from unverified hypotheses and team assumptions.
• Normalizes incident timelines and identifies critical missing data intervals.
• Classifies impact signals based strictly on provided telemetry and reports.
• Generates a list of safe, low-risk checks to disprove leading theories.
• Maps findings to specific specialist roles for efficient handoff.

Why this beats prompting it yourself

General-purpose models often hallucinate root causes or conflate timing with proof. This skill enforces a strict diagnostic discipline that prevents premature conclusions and ensures your next investigation step is based on evidence rather than intuition.

Use cases

• Synthesizing fragmented reports from multiple teams during an active outage.
• Standardizing the first-look brief before escalating to senior SREs.
• Mapping out safe diagnostic steps when deployment timing is known but impact is not.
• Converting raw log dumps and alerts into a structured status report for stakeholders.

Known limitations

This tool performs no external actions and cannot access production environments. It relies entirely on sanitized material provided by the user.