What is the RED method used in this skill?

The RED method is a monitoring strategy that focuses on Request Rate, Errors, and Duration (latency) to measure the health and performance of your services.

Can I use this for distributed systems?

Absolutely. It provides comprehensive patterns for OpenTelemetry distributed tracing and request correlation context propagation across multiple services.

Does this skill support LLM-specific monitoring?

Yes, it includes specialized integration patterns for Langfuse to track LLM tokens, execution costs, and trace complex AI workflows.

What logging formats are prioritized?

The skill prioritizes structured JSON logging, which is machine-parseable and designed for modern log aggregation tools like Loki, ELK, or Datadog.

How does it handle alert fatigue?

It provides strategies for alert grouping, inhibition rules, and severity classification to ensure teams only receive actionable notifications for real issues.

Observability & Monitoring

Name: Observability & Monitoring
Author: yonatangross

byyonatangross

•

Analíticas y Monitorización

Implements comprehensive observability frameworks for structured logging, Prometheus metrics, distributed tracing, and system alerting.

This skill provides production-ready patterns for monitoring application health and performance across the entire stack. It facilitates the implementation of the three pillars of observability—structured logging, Prometheus-compatible metrics, and OpenTelemetry tracing—while also offering specialized monitoring for LLM costs and performance using Langfuse. Whether you are debugging production issues or setting up Service Level Objectives (SLOs), this skill provides the templates and best practices needed to ensure system reliability and operational excellence.

Características Principales

01Standardized health check patterns for Kubernetes liveness and readiness probes

02Distributed tracing setup using OpenTelemetry with auto-instrumentation

03Advanced LLM observability including token usage and cost tracking via Langfuse

04RED method metrics implementation (Rate, Errors, Duration) for Prometheus

05Structured JSON logging with request correlation IDs for cross-service tracking

0669 GitHub stars

Casos de Uso

01Setting up distributed tracing across a microservices architecture to identify latency bottlenecks

02Debugging complex production incidents by correlating logs, metrics, and traces using a unified context

03Implementing custom Prometheus metrics to monitor LLM performance, latency, and operational costs

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add yonatangross/orchestkit observability-monitoring

For use in Claude.ai and ChatGPT

Download Skill