Expert OpenTelemetry guidance for collector configuration, pipeline design, and production telemetry instrumentation across Kubernetes, ECS, serverless, and standalone deployments. Use when configuring collectors, designing pipelines, instrumenting applications, implementing sampling, managing cardinality, securing telemetry, writing OTTL transformations, or setting up AI coding agent observability (Claude Code, Codex, Gemini CLI, GitHub Copilot).
94
98%
Does it follow best practices?
Impact
94%
1.34xAverage score across 18 eval scenarios
Passed
No known issues
{
"context": "Tests whether the skill causes the agent to correctly configure a Kubernetes Deployment (Gateway pattern) for centralized telemetry collection. Unlike DaemonSets, Gateways require load balancing, resilience configuration, and proper service exposure.",
"type": "weighted_checklist",
"checklist": [
{
"name": "Uses Deployment resource type with configurable replicas",
"description": "The response specifies Deployment as the Kubernetes resource type with a replicas field (not fixed to 1). This enables horizontal scaling. Score zero if DaemonSet is suggested instead.",
"max_score": 20
},
{
"name": "Configures LoadBalancer or ClusterIP Service",
"description": "The configuration includes a Service resource (LoadBalancer for external traffic or ClusterIP for internal). Service must be properly labeled and port names must match collector port naming conventions.",
"max_score": 15
},
{
"name": "Includes resource requests, limits, and HPA autoscaling",
"description": "The configuration includes both resource requests/limits and optionally a HorizontalPodAutoscaler for auto-scaling based on CPU/memory. Full score with HPA, partial credit with just requests/limits.",
"max_score": 20
},
{
"name": "Uses persistent queues for resilience",
"description": "The collector configuration includes file_storage extension and persistence queue settings to survive pod restarts. Must include volume mounts for persistence. Score zero if queues are in-memory only.",
"max_score": 15
},
{
"name": "Implements loadbalancing exporter with trace ID routing",
"description": "For backend systems that require consistent trace routing, the configuration includes a loadbalancing exporter with routing_key: traceID. This enables tail sampling across replicas. Score zero if routing is not addressed.",
"max_score": 20
},
{
"name": "Includes PodDisruptionBudget for high availability",
"description": "The configuration includes a PodDisruptionBudget to ensure minimum replicas during node maintenance. Optional but recommended for production.",
"max_score": 10
}
]
}.claude-plugin
.codex-plugin
.cursor-plugin
.github
scripts
docs
evals
scenario-1
scenario-2
scenario-3
scenario-4
scenario-5
scenario-6
scenario-7
scenario-8
scenario-9
scenario-10
scenario-11
scenario-12
scenario-13
scenario-14
scenario-15
scenario-16
scenario-17
scenario-18
references