Does this skill replace the Langfuse core logic?

No, it acts as a routing layer or bridge to existing helper scripts in the mberto-compound repository and does not re-implement Langfuse's core logic.

How does it handle skill synchronization?

It includes a 'skills-sync' utility that checks your current Claude plugin status and installs or copies any missing Langfuse Codex skills into your local skills directory.

What is the primary purpose of the Langfuse Analyzer Commands skill?

It provides a command wrapper for Langfuse, allowing users to run evaluation and observability tasks using Claude-style slash commands like /agent-eval and /setup-dataset.

Can I use this for prompt optimization?

Yes, the skill includes the /optimize-bootstrap command which interfaces with scripts designed to optimize agent performance and configuration.

Which environment variables are required to use this skill?

You must configure LANGFUSE_PUBLIC_KEY and LANGFUSE_SECRET_KEY. LANGFUSE_HOST is also supported as an optional variable.

Langfuse Analyzer Commands

Name: Langfuse Analyzer Commands
Author: mberto10

bymberto10

•

分析与监控

Facilitates LLM agent evaluation and observability by bridging Claude-style commands to Langfuse infrastructure and dataset management.

Langfuse Analyzer Commands acts as a specialized bridge between Claude Code and the Langfuse observability platform, providing a seamless command-line interface for managing LLM evaluation workflows. It enables developers to initialize evaluation infrastructure, set up datasets, and run agent evaluations directly via familiar slash-commands like /agent-eval and /setup-dataset. By providing parity between Codex skills and Claude plugins, this tool simplifies the process of monitoring AI performance, optimizing prompts, and maintaining consistent observability across different development environments without re-implementing core Langfuse logic.

主要功能

01Automated infrastructure initialization and dataset setup scripts

02Seamless synchronization between Claude plugins and Codex skills

03Support for agent evaluation and bootstrap optimization routines

04Unified slash-command interface for Langfuse evaluation workflows

051 GitHub stars

06Built-in status checking for skill parity and installation states

使用场景

01Syncing Langfuse-specific skills across different Claude and Codex environments

02Setting up observability infrastructure for production-grade LLM applications

03Evaluating AI agent performance and accuracy using Langfuse datasets

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add mberto10/mberto-compound langfuse-analyzer-commands

For use in Claude.ai and ChatGPT

Download Skill