Monitors LLM observability, manages prompt versioning, and handles evaluation datasets through the Langfuse platform.
The Langfuse CLI skill empowers Claude to interact directly with the Langfuse LLM observability platform to monitor application performance and manage production prompts. It enables real-time querying of traces, sessions, and metrics like latency and token costs, while also providing robust tools for prompt management and dataset evaluation. Whether you are debugging specific LLM calls, tracking production costs, or deploying new prompt versions with labels, this skill bridges the gap between local development and production monitoring by bringing observability data directly into your workflow.
Características Principales
01Analyze LLM costs, token usage, and latency with granular metrics
02Export data in multiple formats including JSON, CSV, and Markdown
03Manage prompts with version control, labels, and production deployment
040 GitHub stars
05Query and filter LLM traces and sessions for deep observability
06Create and manage evaluation datasets and runs for model testing
Casos de Uso
01Investigating production errors by retrieving detailed trace observations and scores
02Versioning and deploying new prompt templates with labels like 'production' or 'staging'
03Tracking and optimizing LLM API expenditures across different models and environments
