How does this skill handle high-volume LLM traffic?

It provides patterns for buffer-based ingestion using message queues like SQS or Kafka to decouple the application from the ingestion API and prevent data loss.

What is the Langfuse Reference Architecture skill?

It is a specialized capability for Claude Code that provides production-ready blueprints and code implementations for deploying Langfuse LLM observability at scale.

Can this skill help with self-hosted Langfuse deployments?

Yes, it includes specific architecture patterns for self-hosted VPC setups using PostgreSQL RDS and internal load balancing for high availability.

Does it include code for context propagation?

Yes, the skill includes TypeScript implementations using AsyncLocalStorage and Express middleware to ensure trace IDs are propagated throughout the request lifecycle.

Langfuse Reference Architecture

Name: Langfuse Reference Architecture
Author: jeremylongshore

byjeremylongshore

•

1,243

Analytics & Monitoring

Implements production-ready LLM observability infrastructure using standardized Langfuse architecture patterns and best practices.

About

This skill provides expert guidance for architecting and deploying Langfuse, the open-source LLM engineering platform, within production environments. It offers detailed patterns for cloud, self-hosted, and high-scale setups, including specialized implementations for singleton SDK management, asynchronous trace context propagation, and queue-based ingestion to ensure reliable observability for enterprise-grade AI applications. Whether you are setting up basic cloud monitoring or building a buffered, high-throughput tracing system, this skill provides the code patterns and structural blueprints necessary for success.

Key Features

Singleton SDK implementation for efficient resource management
Multi-environment configuration strategies for Dev, Staging, and Production
Production-grade architecture patterns (Cloud, Self-Hosted, High-Scale)
AsyncLocalStorage-based trace context propagation for Node.js
Buffer-based ingestion using SQS/Kafka for high-volume traffic
1,243 GitHub stars

Use Cases

Implementing reliable distributed tracing across microservices
Optimizing Langfuse deployment for high-throughput production environments
Designing LLM observability infrastructure for enterprise AI applications

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add jeremylongshore/claude-code-plugins-plus-skills langfuse-reference-architecture

For use in Claude.ai and ChatGPT

Download Skill

GitHub