Check Pod Resource Usage

Preconditions

Before applying this skill, verify:

• Metrics server is available
• Need to understand pod resource consumption

Actions

1. Get Resource Usage for Pods

Retrieve CPU and memory usage metrics for pods in the namespace.

mcp_tool: kubernetes-mcp-server/pods_top
params:
  namespace: $namespace
timeout: 30s

Success Criteria

The skill succeeds when:

• Pod resource metrics retrieved successfully
• High resource consumers identified if present

Failure Handling

If metrics unavailable:

1. Check if metrics-server is deployed
2. Fall back to resource requests/limits from pod spec

Examples

Input Context:

{
  "namespace": "production",
  "reason": "investigating high memory usage"
}

Expected Outcome: Pod metrics retrieved showing CPU and memory usage for each pod, with any pods consuming excessive resources clearly identified.