What authentication is required for Langfuse?

You must set the LANGFUSE_PUBLIC_KEY and LANGFUSE_SECRET_KEY environment variables to connect the skill to your Langfuse project.

Does updating a prompt overwrite the previous one?

No, updating a prompt automatically creates a new version in Langfuse, preserving your full history for easy rollbacks.

Does this skill support chat-based prompts?

Yes, it supports both simple text prompts and structured JSON chat prompts including system, user, and assistant roles.

Can I see what changed between two prompt versions?

Yes, the compare command provides a diff-style output that highlights changes in the prompt text, chat structure, or configuration settings.

How do I deploy a prompt to production?

You can use the promote operation to assign the 'production' label to any specific prompt version, making it the default for your application.

Langfuse Prompt Management

Name: Langfuse Prompt Management
Author: mberto10

bymberto10

0•

Analytics & Monitoring

Manages the complete lifecycle of Langfuse LLM prompts, including version control, deployment labels, and side-by-side version comparisons.

The Langfuse Prompt Management skill provides a powerful interface for developers to handle prompt engineering workflows directly within Claude Code. It enables full CRUD operations for both text and chat-based prompts stored in Langfuse, allowing users to create new versions, assign deployment labels like 'production' or 'staging', and compare different iterations via diff-style outputs. This skill bridges the gap between prompt development and production deployment, ensuring that your LLM application templates are versioned, tested, and promoted through a structured workflow.

Key Features

01Side-by-side version comparison with diff-style output

02Full CRUD operations for text and chat-type Langfuse prompts

03Version control system with custom commit messages

04Environment management using deployment labels (Production/Staging/Latest)

050 GitHub stars

06Support for model configuration objects and metadata storage

Use Cases

01Promoting a validated prompt version from a staging environment to production

02Debugging prompt behavior by comparing the differences between two specific versions

03Iterating on LLM prompt templates and tracking changes through version history

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add mberto10/mberto-compound prompt-management

For use in Claude.ai and ChatGPT

Key Features

01Side-by-side version comparison with diff-style output

02Full CRUD operations for text and chat-type Langfuse prompts

03Version control system with custom commit messages

04Environment management using deployment labels (Production/Staging/Latest)

050 GitHub stars

06Support for model configuration objects and metadata storage

Use Cases

01Promoting a validated prompt version from a staging environment to production

02Debugging prompt behavior by comparing the differences between two specific versions

03Iterating on LLM prompt templates and tracking changes through version history

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add mberto10/mberto-compound prompt-management

For use in Claude.ai and ChatGPT