What problem does it solve? Node.js services often run blind in production: slow Prisma queries, failing BullMQ jobs, and scattered logs with no way to connect them to a single request. This Skill generates a complete OpenTelemetry setup so every request, query, and job is traced, measured, and logged with correlation IDs. ## Core Features & Use Cases - SDK Initialization & Auto-Instrumentation: Generates instrumentation.ts with NodeSDK, OTLP trace and metric exporters, and auto-instrumentation for Fastify, Prisma, and HTTP, enforcing the critical rule that OTel loads before all other imports. - Custom Spans & Queue Propagation: Wraps business logic in named spans with attributes and error recording, and propagates trace context from BullMQ producers to workers. - RED Metrics & Structured Logging: Configures rate/error/duration histograms, observable gauges for active jobs, and a Pino logger that injects traceId and spanId into every log line. - Use Case: You have a Fastify API backed by Prisma and BullMQ and cannot tell why requests are slow in production. This Skill produces the instrumentation file, entry-point wiring, metrics, logger, and a Docker Compose stack with Jaeger and Prometheus so you can see every slow query. ## Quick Start Ask the AI to add OpenTelemetry tracing, RED metrics, and trace-correlated logging to your Fastify and Prisma service, exporting to your preferred backend such as SigNoz, Grafana, Jaeger, or Datadog.
