Troubleshoot AWS and EKS
Use for a concrete incident, error, or broken cloud behavior.
Workflow
Triage scope, impact, severity, and recent changes; collect evidence; form and test a hypothesis; apply authorized mitigation; reproduce the original operation; document cause, recovery, and prevention.
Coverage
Pod scheduling/crash/image/OOM failures; node conditions and kubelet; API/add-on errors; CNI/DNS/load-balancer paths; PVC/CSI topology; IAM and RBAC denials.
Reporting and boundaries
P1/P2 incidents or symptoms spanning multiple domains can use the coordinator. Preserve timestamps and correlation evidence; do not infer one root cause merely because symptoms occurred together.
The source skill contains the command playbooks, decision trees, examples, reference files, and host/team integration. Consult it for the exact operation being performed.