What problem does it solve? New API endpoints and services often ship without any logging, metrics, or alerting, leaving teams blind during the first production incident. This Skill instruments a system for production at endpoint creation time, before anything is deployed. ## Core Features & Use Cases - Structured Logging Setup: Generates OpenTelemetry-compliant JSON logging configuration with the six mandatory fields (timestamp, severity, trace_id, span_id, service.name, body) for Python, Go, TypeScript, Java, and Rust. - Golden Signal Metrics: Instruments latency histograms, request counters, error counters, and saturation gauges per endpoint using the OpenTelemetry SDK. - SLO, Alert, and Runbook Generation: Produces an SLO definition document with error budget policy, symptom-based alert rules (including burn-rate alerts), and a runbook per alert. - Use Case: After creating a new checkout API endpoint, run this Skill to generate structured logging config, Prometheus alert rules, an SLO document with a 99.9% availability target, and runbooks for high-error-rate and high-latency alerts, then validate everything with the observability-reviewer agent. ## Quick Start Ask the AI to instrument the newly created API endpoints with structured logging, golden signal metrics, SLOs, alert rules, and runbooks before deployment.
