Most tools target a distinct resource+action (list_pods vs describe_pod vs get_pod_logs, namespace events vs namespace health), so selection is generally clear. The main overlap is between debug_dns_config and coredns_health_check, which both cover CoreDNS pod health and could be confused, and the service trio (find_orphaned_services, check_service_endpoints, trace_service_to_pods) sits close together though descriptions differentiate them.

17 of 19 tools follow a verb_noun pattern (list_pods, get_pod_logs, describe_deployment, check_service_endpoints, trace_service_to_pods). Only coredns_health_check and node_network_status break the pattern with noun-first phrasing, a minor deviation that stays readable.

19 tools covers a genuinely broad surface (pods, deployments, events, DNS, network policy, ingress, mesh, Dynatrace), so each largely earns its place. It sits at the heavy end of the comfortable range, and the near-duplicate CoreDNS pair is the one spot that looks like padding.

The observability/network-debugging niche is well covered: pods, deployments, logs, events, health rollups, DNS, ingress, mesh, and Dynatrace correlation. Noticeable gaps are a plain list_services, basic node listing beyond node_network_status, and resource-usage metrics (CPU/memory) that agents would expect from an observability server, though most are workable around via describe commands.