monitoring-stack-deployer

What problem does it solve?

Many teams struggle to reliably deploy a production-grade monitoring and alerting stack that provides accurate, low-noise signals tied to user impact. This Skill streamlines provisioning Prometheus, Grafana, ServiceMonitors, recording rules, and SLO-driven alerting so monitoring is scalable, maintainable, and actionable.

Core Features & Use Cases

• SLO-driven alerting: Configure burn-rate alerts and multi-window policies to reduce alert fatigue and align notifications with user impact.
• Kubernetes-native discovery: Deploy kube-prometheus-stack and use ServiceMonitor/PodMonitor to automatically discover application metrics endpoints.
• Recording rules & performance: Precompute expensive queries with Prometheus recording rules to improve alert evaluation speed and reliability.
• Dashboard-as-code: Provision Grafana dashboards (RED for services, USE for infrastructure) and expose dashboards as code for reproducible observability.
• Operational patterns: Includes guidance on alert routing, runbooks, retention policy, and long-term storage integrations (Mimir/Thanos/VictoriaMetrics).

Quick Start

Set up kube-prometheus-stack in your Kubernetes cluster, create ServiceMonitors for your services, define SLOs and recording rules, and provision a Grafana dashboard that surfaces RED metrics and SLO burn-rate alerts.