Monitoring Setup

Monitoring Setup is a skill that generates complete observability configurations for your infrastructure and applications. It covers the full stack from metric collection to dashboards to alerting, following Google SRE and OpenTelemetry best practices.

How It Works

1. Stack assessment — Identifies what you're running and recommends the right monitoring approach
2. Metric collection — Generates Prometheus scrape configs, recording rules, and service discovery
3. Dashboard creation — Produces Grafana dashboard JSON with RED metrics (Rate, Errors, Duration)
4. Alert design — Creates multi-level alert rules with proper severity, grouping, and inhibition
5. SLO definition — Helps define Service Level Objectives with error budget burn rate alerts

Best For

• Setting up monitoring for new Kubernetes clusters or services
• Migrating from legacy monitoring to Prometheus/Grafana stack
• Implementing SLO-based alerting to reduce alert fatigue
• Adding distributed tracing with OpenTelemetry

Philosophy

Follows the Google SRE approach: monitor symptoms not causes, alert on SLO violations not individual metrics, and use dashboards for diagnosis not detection.