k8s-ops

Troubleshoot Kubernetes CrashLoopBackOff pods by checking pod status, viewing error logs, and inspecting events across EKS, GKE, and AKS clusters. You can safely diagnose issues without modifying your global kubectl context or risking unintended production changes.

You can scale deployments and rollout restart workloads on production EKS clusters, but the skill enforces mandatory user confirmation for all change operations. This safety guardrail prevents unintended modifications to your production environments.

Manage multi-cluster Kubernetes resources without changing kubectl context by using context-specific cluster targeting. View the status of Pods, Deployments, and Services across AWS EKS, GCP GKE, and Azure AKS clusters while avoiding global state modifications.

Exec into containers and set up port-forwarding to diagnose network connectivity issues directly. Access container shells and forward local ports to services across your managed Kubernetes clusters for rapid troubleshooting.

Applying Kubernetes configuration changes requires manual confirmation for every operation. This mandatory approval workflow applies to all modifications including scaling, rollout restarts, and configuration applies to prevent accidental changes.

High-risk actions are prohibited on production Kubernetes clusters during workload operations to prevent accidental production outages. The skill enforces safety guardrails that restrict dangerous modifications in production environments.