Safely evict all pods from a node for maintenance. Cordons the node first, then evicts pods respecting PodDisruptionBudgets. Use before node maintenance or decommissioning. Keywords: drain, evict, node maintenance, pod migration.
Deploy and operate AI workloads on Kubernetes with GPU scheduling, model serving, and MLOps patterns
Diagnose and fix issues with the Kubernetes cluster, Flux reconciliation, or deployed applications
Mayastor block storage and SeaweedFS object storage patterns for Kubernetes persistent data.
Troubleshooting workflows for OpenShift 4.20 homelab. Diagnose PVC pending issues, pod crashes, CSI driver problems, storage network connectivity (VLAN 160), and operator failures. Use when debugging cluster issues or investigating stuck resources.
Kubernetes cost management, resource optimization, and FinOps practices
Set up multi-container applications with Docker Compose including services, networks, and volumes
Docker イメージを管理する。「イメージ一覧」「docker images」「イメージ確認」「ビルドして」「docker build」「イメージ作成」「イメージ取得」「docker pull」「イメージダウンロード」「イメージの履歴」「docker history」などで起動。
Scale a deployment to a specified number of replicas. Used for both scaling up (to handle load) and scaling down. Keywords: scale, replicas, deployment, statefulset, horizontal scaling, pod count.
Implement persistent storage with Docker volumes, bind mounts, and backup strategies
Bootstrap a new node for Kubernetes without joining it to the cluster. Use when preparing nodes in advance, testing setup, or staging hardware.
Set a key and value in the keyvalue storage. Requires authentication. Use for Agentuity cloud platform operations
Use when the user asks for Kubernetes cluster inspection and troubleshooting (kubectl + kustomize; non-destructive by default).
Diagnoses and resolves issues with Flux GitOps deployments, Kubernetes pods, services, and HelmReleases in the Superbloom cluster
Docker containerization best practices. Use when building Docker images, writing Dockerfiles, configuring Docker Compose, or troubleshooting container issues.
Deploy and manage hybrid cloud media applications with rclone sidecar pattern. Use for adding media apps (Plex, Sonarr, Radarr, etc.), implementing cloud storage mounts (Zurg/TorBox), troubleshooting FUSE mount propagation, and checking media mount status across pods.
Diagnose storage and persistent volume issues. Checks PVC binding, storage class configuration, and volume mount status. Use when pods are stuck pending due to volume issues. Keywords: PVC, PV, storage, volume, pending, mount, storage class.
Apply when managing application configuration: environment variables, secrets management, and config validation.
Collect recent Kubernetes events to identify issues. Events reveal warnings, errors, and state changes. Keywords: events, warnings, errors, cluster activity, state changes, recent events.
Helm chart development workflow including chart structure, values management, testing, linting, and publishing for EKS deployments with Keycloak integration
Deploy F# applications with Tailscale sidecar for private network access. Use when: "deploy", "production", "Tailscale", "docker-compose", "Docker", "Portainer", "private network", "deployment", "ship it", "go live", "home server", "self-host", "container", "release". Creates docker-compose.yml with app + Tailscale sidecar for secure access. No public ports, no authentication needed - Tailscale handles it.
