incident-root-cause-analyzer

What problem does it solve?

This Skill provides a comprehensive, end-to-end methodology for analyzing production incidents in distributed systems, enabling teams to quickly identify cascading failures, resource contention, and backpressure-driven root causes. It helps reduce mean time to detection and resolution by automating data collection, anomaly scoring, hypothesis testing, and evidence-based reporting.

Core Features & Use Cases

• Automated root-cause analysis across microservice boundaries using logs, metrics, and traces.
• Anomaly detection with timeline correlation to reveal the sequence of events leading to an incident.
• Evidence-based hypothesis testing and automated visualizations (Mermaid diagrams) to communicate findings.
• Generates formal Root Cause Analysis reports and structured evidence artifacts to speed incident reviews.

Quick Start

Load the incident root cause analyzer skill into Claude, provide:

• Incident time window (e.g., 07:00:00 ± 30s)
• Metrics data directory (CSV files)
• Optional logs or traces directory Then run the analysis to produce a Root Cause Analysis report, evidence charts, a fault evolution diagram, and an evidence index.