observability-integrate

To collect Prometheus metrics, add ServiceMonitor and PodMonitor definitions co-located with your app manifests. This skill wires them into kube-prometheus-stack, applying monitors to enable consistent scrape target collection across namespaces without manual per-app wiring.

Create Grafana dashboards as ConfigMaps using the sidecar pattern for kube-prometheus-stack. This skill generates sidecar-configured ConfigMaps that automatically load into Grafana, providing immediate visibility of app health metrics scraped by Prometheus.

Yes, this integration assumes Flux-based repo patterns with manifests co-located in kubernetes/apps/<ns>/<app>/app. It validates monitors, dashboards, and alert rules against ADR 0007 and plan 0002, requiring this GitOps structure to function correctly.

Prometheus may not be scraping your app because the ServiceMonitor configuration is misapplied. This skill verifies scrape targets and diagnoses missing metrics using prescribed probe recipes, validating that monitors are correctly wired into kube-prometheus-stack.

Set up PrometheusRule alerts by authoring rule groups for recording and alerting conditions. This skill applies PrometheusRule definitions alongside ServiceMonitors and Grafana dashboards, validating alert configurations within the kube-prometheus-stack observability stack.

Yes, you can debug missing metrics and verify scrape targets using this skill. It diagnoses metric visibility issues by applying prescribed probe recipes that validate Prometheus scrapes and dashboard integrity across your observability stack configuration.