Build Production-Ready Observability

Most services go to production with a "black box" mentality, leading to frantic firefighting during outages. This skill provides a disciplined, developer-centric framework for instrumenting any service with high-fidelity distributed tracing, structured logging, health probes, and RED metrics.

What it does

Unlike basic prompts, this skill follows a rigorous multi-phase process to assess your architecture and implement the four pillars of observability:

• Tracing: Root and child spans with W3C context propagation.
• Structured Logging: JSON-formatted logs correlated with Trace IDs for instant debugging.
• Health Probes: Kubernetes-ready liveness, readiness, and startup checks with dependency validation.
• RED Metrics: High-signal metrics (Rate, Errors, Duration) without the risk of high-cardinality explosions.

Supported Tools & Frameworks

The skill is designed for the modern cloud-native stack, recommending OpenTelemetry (OTel) by default while supporting vendor-specific SDKs like Datadog, New Relic, and AWS X-Ray. It works across Node.js, Python, Go, and Java, providing boilerplate and configuration tailored to your specific deployment environment (Kubernetes, ECS, Lambda, etc.).

Why use this skill

Manual instrumentation is error-prone—it's easy to leak PII in logs or forget to propagate trace headers across service boundaries. This skill ensures a standardized implementation that adheres to industry best practices, giving you a cohesive dashboard-ready output from day one.