Auto-discovered marketplace from kanr/claude-plugin-kubernetes
npx claudepluginhub kanr/claude-plugin-kubernetesKubernetes cluster awareness, diagnostics, and remediation via kubectl. Adds 21 tools for inspecting pods, deployments, services, events, logs, and performing safe cluster operations.
RuFlo Marketplace: Claude Code native agents, swarms, workers, and MCP tools for continuous software engineering
Claude Code marketplace entries for the plugin-safe Antigravity Awesome Skills library and its compatible editorial bundles.
Production-ready workflow orchestration with 80 focused plugins, 185 specialized agents, and 153 skills - optimized for granular installation and minimal token usage
Share bugs, ideas, or general feedback.
An MCP server that gives Claude deep Kubernetes cluster awareness, diagnostics, and remediation capabilities via kubectl.
| Tool | Description |
|---|---|
k8s_cluster_info | Current context, server version, API endpoint |
k8s_get_contexts | List all kubeconfig contexts, indicate active one |
k8s_list_namespaces | List namespaces with status and age |
k8s_list_nodes | Nodes with roles, status, version, OS, IP, age |
k8s_list_pods | Pods with status, restarts, node (filter by ns/label) |
k8s_list_deployments | Deployments with replica counts and age |
k8s_list_services | Services with type, cluster IP, external IP, ports |
k8s_list_images | Container images running across pods |
k8s_list_events | Cluster events sorted by time (filterable by Warning type) |
k8s_list_statefulsets | StatefulSets with desired/ready counts and age |
k8s_list_ingresses | Ingress resources with hosts, paths, backends |
k8s_list_jobs | Jobs with completions, duration, age |
k8s_list_cronjobs | CronJobs with schedule, last schedule time, active count |
k8s_list_configmaps | ConfigMaps (metadata only) |
k8s_list_secrets | Secrets (metadata only — never exposes data) |
k8s_list_pvcs | PVCs with status, volume, capacity, access modes |
k8s_list_daemonsets | DaemonSets with desired/current/ready counts |
k8s_list_hpa | HPA with target, min/max replicas, metrics |
k8s_list_networkpolicies | NetworkPolicies with pod selector and age |
k8s_api_resources | Available API resource types including CRDs |
| Tool | Description |
|---|---|
k8s_describe | kubectl describe any resource |
k8s_logs | Pod logs with tail, container, previous, since, and error filtering |
k8s_top_pods | Pod CPU/memory usage (requires metrics-server) |
k8s_top_nodes | Node CPU/memory usage (requires metrics-server) |
k8s_find_issues | Comprehensive health scan — pods, nodes, deployments, statefulsets, jobs, PVCs, events |
k8s_get_yaml | Resource as YAML (managed fields stripped by default; raw=true for full output) |
k8s_self_test | Plugin health check — kubectl binary, cluster connectivity, auth, metrics-server |
| Tool | Risk | Description |
|---|---|---|
k8s_restart_deployment | Low | Rolling restart a deployment |
k8s_scale | Medium | Scale deployment/statefulset (scale-to-zero requires explicit confirmation) |
k8s_delete_pod | Low-Med | Delete pod for controller recreation; force=true for stuck pods |
k8s_rollback_deployment | Medium | Roll back to previous or specific revision |
k8s_apply_manifest | Medium | Apply YAML/JSON manifest; dry-run supported; cluster-scoped resources blocked by default |
k8s_patch_resource | Medium | Strategic merge, JSON merge, or JSON patch any resource |
k8s_node_operation | High | Cordon / uncordon / drain a node |
k8s_rollout_status | Read-only | Check rollout progress (30s timeout) |
k8s_rollout_history | Read-only | View all recorded revisions for a deployment |
k8s_delete_resource | Medium | Delete any resource by type and name |
k8s_diff | Read-only | Diff a manifest against live cluster state |
Static resources and namespace-scoped templates for direct data access:
| URI | Description |
|---|---|
k8s://contexts | List kubeconfig contexts |
k8s://cluster-info | Current context, server version, API endpoint |
k8s://namespaces | All namespaces with status and age |
k8s://namespaces/{namespace}/pods | Pods in a namespace |
k8s://namespaces/{namespace}/deployments | Deployments in a namespace |
k8s://namespaces/{namespace}/services | Services in a namespace |
k8s://namespaces/{namespace}/events | Events in a namespace |
Structured troubleshooting workflows:
| Prompt | Description |
|---|---|
diagnose-pod | Step-by-step pod diagnosis (describe, events, logs, resources) |
cluster-health-report | Full health scan with drill-down into issues |
incident-response | Incident triage — scan, events, failing pods, blast radius |
debug-crashloop | CrashLoopBackOff debugging — previous logs, OOM check, probes |
pre-deploy-checklist | Pre-deployment verification — deployments, pods, resources, nodes |
Add to your project's .mcp.json:
{
"mcpServers": {
"kubernetes": {
"command": "uvx",
"args": ["--from", "claude-plugin-kubernetes", "k8s-mcp"]
}
}
}
To verify, run /mcp inside Claude Code. You should see kubernetes listed with 38 tools.
Add to ~/Library/Application Support/Claude/claude_desktop_config.json (macOS) or %APPDATA%\Claude\claude_desktop_config.json (Windows):