design-monitoring

What problem does it solve?

Many projects have monitoring specifications but lack an automated, repeatable way to turn those specs into concrete observability artifacts and verify runtime emissions; this Skill implements specs from augur into Grafana dashboards, Prometheus alert rules, and live metric validation to close that gap.

Core Features & Use Cases

• Spec-driven generation: Parse augur's monitoring-spec.yaml and atlas context to generate project-specific Grafana overview and vitals dashboards and Prometheus alert rule groups.
• Alert and meta-alert creation: Produce standard alert rule files with severity labels, annotations, and a required VitalsMissing meta-alert.
• Runtime validation: Optionally access the cluster to scrape /metrics, confirm metric names/types/labels, verify infra-atlas contract requirements, and produce a validation report classifying PASS/MISSING/TYPE_MISMATCH/LABELS_MISSING/CONTRACT_VIOLATION.
• Memory-first outputs: Write generated dashboards, alerts, and implementation/validation reports to the sauron project memory for review, and optionally push dashboards and apply rules when not in dry-run.
• Use Case: Given augur's approved monitoring-spec for a Kubernetes service, produce ready-to-provision Grafana JSON, Prometheus rules, and a validation report to onboard observability quickly.

Quick Start

Generate monitoring for my-project from the monitoring-spec.yaml and write dashboards, alerts, and a validation report to the sauron project memory as a dry run.