get-cluster-health
Get overall cluster health status. Checks node readiness, system pod health, and control plane components. Use for periodic health checks or incident investigation. Keywords: cluster health, node status, cluster status, health check.
When & Why to Use This Skill
This Claude skill provides a comprehensive health check for Kubernetes clusters, automatically assessing node readiness, system pod status, and critical cluster events. It streamlines infrastructure monitoring and incident investigation by delivering a real-time snapshot of cluster stability and performance, making it an essential tool for SREs and DevOps engineers.
Use Cases
- Routine Infrastructure Monitoring: Perform scheduled health checks to ensure all nodes and system components are functioning optimally and meet high-availability standards.
- Rapid Incident Triage: Quickly identify if service disruptions are caused by underlying cluster-level issues such as node failures, NotReady states, or system pod crashes.
- Post-Deployment & Maintenance Validation: Verify cluster stability immediately after performing upgrades, configuration changes, or scaling operations to ensure no regressions occurred.
- Automated SRE Diagnostic Reporting: Generate instant, structured status reports on cluster health for on-call handovers or periodic system audits.
| name | get-cluster-health |
|---|---|
| description | > |
| investigation. Keywords | cluster health, node status, cluster status, health check. |
| domain | k8s |
| category | collection |
| requires-approval | false |
| confidence | 0.9 |
Get Cluster Health
Preconditions
Before applying this skill, verify:
- Kubernetes cluster is accessible
- kubectl/MCP has cluster-admin or read permissions
Actions
1. Get Node Status
List all nodes and their readiness status.
mcp_tool: kubernetes-mcp-server/resources_list
params:
apiVersion: v1
kind: Node
timeout: 30s
2. Get System Pod Health
Check pods in kube-system namespace for any issues.
mcp_tool: kubernetes-mcp-server/pods_list_in_namespace
params:
namespace: kube-system
timeout: 30s
3. Get Recent Events
Check for warning events across the cluster.
mcp_tool: kubernetes-mcp-server/events_list
params: {}
timeout: 30s
Success Criteria
The skill succeeds when:
- All nodes are in Ready state
- All kube-system pods are Running
- No critical warning events in last hour
Failure Handling
If nodes are NotReady or pods are failing:
- Capture current state for analysis
- Trigger diagnostic skills for specific issues
- Escalate if multiple nodes or critical pods affected
Examples
Output:
{
"nodes": {
"total": 3,
"ready": 3,
"not_ready": 0
},
"system_pods": {
"total": 12,
"running": 12,
"failed": 0
},
"warnings": 0,
"healthy": true
}