skills.rest
ajbmachon
ajbmachonCOMMUNITY
Explore all skills by this author →
Website𝕏@ajbmachon2GitHub

langfuse

Analyze Langfuse traces, errors, costs, and evaluation results to diagnose AI issues.

7|2|Updated Nov 30, 2025
Search Tags:#evaluation#experimentation#traces#langfuse#error investigation#cost monitoring#latency analysis
One-click install
npx skills add https://github.com/ajbmachon/ajbm-skills --skill langfuse-ajbmachon
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: langfuse
Source: https://github.com/ajbmachon/ajbm-skills/tree/main/.claude/skills/langfuse
Command: npx skills add https://github.com/ajbmachon/ajbm-skills --skill langfuse-ajbmachon

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires uv, python, langfuse, python-dotenv, and includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill eliminates guesswork in AI ops by turning raw Langfuse traces, errors, and evaluation data into evidence-backed insights and prioritized next actions.

Core Features & Use Cases

  • Trace forensics: list, fetch, and analyze traces to pinpoint latency bottlenecks and failure points.
  • Evaluation & scoring: design evaluation strategies and create score records from specific traces.
  • Experiment workflows: create datasets, add items, run experiments, and compare runs using Langfuse’s evaluation primitives.

Quick Start

Ask: "Analyze my Langfuse trace <trace-id> and summarize the top latency bottlenecks and any errors, then recommend the next validation commands."

Frequently Asked Questions about langfuse

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I analyze Langfuse traces to pinpoint latency bottlenecks and failures?▼

Analyzing Langfuse traces involves listing and fetching trace records to pinpoint latency bottlenecks and failure points. This diagnostic process transforms raw telemetry into evidence-backed summaries with specific next commands.

How do I design evaluations and score traces using Langfuse?▼

Designing Langfuse evaluations involves creating evaluation strategies and generating score records directly from specific traces. This allows you to systematically assess and score AI application quality based on trace data.

Can I run dataset-based experiments and A/B testing with Langfuse?▼

Yes, you can run dataset-based experiments for regression and A/B testing using Langfuse. This workflow supports creating datasets, adding items, running experiments, and comparing runs using evaluation primitives.

Do I need specific environment variables to monitor AI costs and errors with Langfuse?▼

Yes, Langfuse cost monitoring and error investigation require Langfuse credentials configured as environment variables. This setup enables the local CLI wrapper script to access traces and produce evidence-backed insights.

What is the best way to debug AI application latency and failures using traces?▼

The best way to debug AI application latency and failures is through trace forensics, which lists, fetches, and analyzes traces to pinpoint exact bottlenecks. This eliminates guesswork by providing evidence-backed diagnostic summaries.

Related Skills

Explore More Data & Analytics Skills→
digital-stoic-orgdigital-stoic-org
19

install-dependency

Automate monorepo-aware dependency installation across Python, JavaScript, and system packages.

Official
Intermediate
ethan-rioethan-rio

arize-experiment

Create, run, and analyze Arize experiments against versioned datasets.

Community
Advanced
jxtngxjxtngx
2

langsmith-dataset

Convert LangSmith trace logs into structured JSON evaluation datasets.

Community
Intermediate
giosuetedeschi-specgiosuetedeschi-spec

langfuse

Track LLM traces, spans, and metrics for production AI applications.

Community
Advanced