What problem does it solve? Production outages and system failures demand fast, structured diagnosis, but engineers often waste time guessing at root causes without a systematic approach. This Skill provides a disciplined incident response methodology covering log analysis, distributed tracing, Kubernetes debugging, and performance analysis to resolve issues quickly and prevent recurrence. ## Core Features & Use Cases - Observability-Driven Debugging: Analyze logs, metrics, and distributed traces across platforms like Prometheus, Grafana, ELK, Jaeger, and OpenTelemetry to form evidence-based hypotheses. - Kubernetes & Container Troubleshooting: Debug OOMKilled pods, CNI networking issues, service mesh traffic problems, and persistent volume failures with kubectl-centric workflows. - Root Cause Analysis & Prevention: Conduct blameless postmortems, document findings, and add proactive monitoring and alerting to stop recurring incidents. - Use Case: Your production API returns intermittent 504 gateway timeouts behind a load balancer. Use this Skill to correlate load balancer logs, upstream service traces, and database connection pool metrics to isolate the bottleneck and implement a fix. ## Quick Start Ask the assistant to debug a specific production issue, for example: "Debug high memory usage in my Kubernetes pods causing frequent OOMKills and restarts."
