| name | setup-monitoring |
| description | Guide for setting up or extending monitoring (metrics, logs, alerts) for a service. Use when the user asks to set up monitoring or add alerts. |
Skill: Setup monitoring
Trigger
When the user asks to set up monitoring, add alerts, or configure dashboards.
Steps
- Scope — Identify what to monitor (API latency, errors, DB, queue depth) and where (cloud provider, vendor).
- Metrics — Define 3–5 key metrics (e.g. request rate, error rate, p99 latency); use structured logging with correlationId.
- Alerts — Propose thresholds (e.g. error rate > 1%, p99 > 2s) and severity; document in runbook.
- Dashboard — List panels to add (time series, status); use placeholders for dashboard URLs.
- Logs — Remind: no PII or secrets; JSON format; include correlationId per project rules.
Rules
- Follow
.cursor/rules/devops/monitoring.mdc. Use placeholder resource names.