langfuse

Analyze Langfuse traces, prompts, datasets, scores, and sessions via CLI.

2|Updated Jun 25, 2026
One-click install
npx skills add https://github.com/cc-claws/cc-code --skill langfuse-cc-claws
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: langfuse
Source: https://github.com/cc-claws/cc-code/tree/main/.claude/skills/langfuse
Command: npx skills add https://github.com/cc-claws/cc-code --skill langfuse-cc-claws

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill helps you inspect, understand, and act on Langfuse data without manually stitching together trace queries, documentation lookups, and analysis workflows. It turns observability tasks that are usually slow and fragmented into a guided, repeatable process.

Core Features & Use Cases

  • CLI-Based Langfuse Access: Query and manage traces, prompts, datasets, scores, sessions, and other API resources programmatically.
  • Documentation Retrieval: Look up Langfuse concepts, integration guides, SDK usage, and feature behavior directly from the official docs.
  • Trace Analysis Workflows: Run bundled Bun scripts to inspect token usage, cache behavior, tool calls, prompt stability, and context growth.
  • Operational Guidance: Supports cost analysis, error analysis, prompt migration, SDK upgrades, user feedback capture, and judge calibration.
  • Use Case: Use this Skill to diagnose why a trace became expensive, identify repetitive tool calls, or confirm whether a prompt migration is safe before rollout.

Quick Start

Ask the assistant to analyze your Langfuse traces, fetch the relevant documentation, or run the built-in analysis workflow for the specific trace or issue you want to review.

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 identify token usage and repetitive tool calls?

To analyze Langfuse traces for token usage and tool calls, you can run bundled Bun scripts that inspect cache behavior, context growth, and prompt stability. This workflow turns fragmented trace queries into a guided inspection process.

What is the best way to perform cost analysis and error analysis on LLM observability data?

Cost analysis and error analysis on LLM observability data are performed through CLI-driven API access and documentation lookup. This approach programmatically queries scores and sessions to diagnose why a specific trace became expensive.

Do I need authenticated Langfuse host credentials to query prompts and datasets via CLI?

Yes, you need authenticated Langfuse host credentials to query prompts, datasets, and sessions. The Skill requires these credentials alongside CLI, WebFetch, curl, and Bun-based scripts to execute trace inspection.

Can I use this workflow to verify if a prompt migration is safe before rollout?

Yes, you can verify if a prompt migration is safe before rollout by running built-in analysis workflows. These workflows compare prompt stability and inspect trace behavior to confirm migration safety.

How does CLI-based Langfuse access handle user feedback capture and judge calibration?

CLI-based Langfuse access handles user feedback capture and judge calibration by querying API resources programmatically. It manages scores and sessions to support observability workflows and judge calibration operations.

Why are my Langfuse observability tasks slow and fragmented when diagnosing trace issues?

Langfuse observability tasks are often slow and fragmented because they require manually stitching trace queries with documentation lookups. This Skill automates the workflow to generate reports and inspect trace data repeatably.