What problem does it solve?
Distributed polyglot services often lack consistent, correlated observability, making it difficult to debug latency issues, trace request flows across services, and monitor system health without excessive telemetry costs or fragile custom instrumentation.
Core Features & Use Cases
- Cross-Language Instrumentation: Auto and manual setup for Java (including Spring Boot) and Python (including FastAPI) services, with built-in support for context propagation, baggage, and semantic conventions.
- OTel Collector Pipeline Configuration: Pre-built YAML templates for receivers, processors, and exporters to route traces, metrics, and logs to backends like Jaeger, Prometheus, Tempo, and Grafana Cloud.
- Sampling Strategy Design: Head and tail sampling configurations to balance telemetry volume and visibility, including error and latency-based sampling to capture critical production issues without overloading backends.
- Production Guardrails: Checklists and anti-pattern guidance to avoid common pitfalls like span leaks, missing resource attributes, and blocking export processors.
Use Case: For a Kafka-based order processing microservices system, use this skill to instrument all Java and Python services, configure tail sampling to capture 100% of failed transactions, and set up a centralized OTel Collector to export telemetry to your existing observability stack.
Quick Start
Use the opentelemetry-deep-dive skill to instrument your Python FastAPI order service with auto-instrumentation and configure the OTel Collector to export traces to Tempo and metrics to Prometheus.