langfuse: Boost LLM Observability and Prompt Management
From the repo · “🪢 Open source agent evals & observability: Trace, evaluate, and improve LLM applications with one open platform.”
Develops and operates large-language-models with tools for evaluation, observability, and metrics.
"A founder uses langfuse to monitor LLM performance and optimize it for better results. First, they install langfuse and set up OpenTelemetry for metrics and observability. Next, they create a prompt management system and configure it to integrate with the LLM. Finally, they run a series of tests to fine-tune the model's performance. With langfuse, they're able to scale their LLM operations efficiently and make data-driven decisions."
- •When you need to monitor and optimize the performance of your LLMs
- •When you want to manage and version your prompts and datasets for LLMs
- •When you need to integrate your LLMs with other tools and platforms like OpenTelemetry, Langchain, or OpenAI SDK
Run 'npm run eval -- --model <your-llm-model> --prompt <your-prompt-text>' to evaluate your LLM's performance on a specific prompt
Skip the builder — one click puts this in Claude, Cursor, Antigravity and more.
Expect 20–40 minutes in a terminal — or let your AI agent drive it.
mkdir -p ~/.claude/skills/langfuse && curl -fsSL https://workflowstacks.com/api/skills/langfuse/claude-skill -o ~/.claude/skills/langfuse/SKILL.mdOpens the app with this repo with the prompt ready to go — no copy-paste needed.
Langfuse: Boost LLM Observability and Prompt Management is a very large TypeScript project (~585k lines across 4317 code files, plus 444 test files). It is a full software project: use it through its install path rather than reading it end to end. Last commit this month, Other license, has a test suite.
- 1README.mdStart here — what it does and how to install it
- 2AGENTS.mdThe instructions the AI actually follows
- 3package.jsonDependencies and the commands it exposes
Skip the builder — one click puts this in Claude, Cursor, Antigravity and more.
Expect 20–40 minutes in a terminal — or let your AI agent drive it.
mkdir -p ~/.claude/skills/langfuse && curl -fsSL https://workflowstacks.com/api/skills/langfuse/claude-skill -o ~/.claude/skills/langfuse/SKILL.mdOpens the app with this repo with the prompt ready to go — no copy-paste needed.
Are you the creator of this tool? and earn 85% of every sale.
Show it off in your README:
[](https://workflowstacks.com/skills/langfuse?utm_source=github&utm_medium=badge)The five fastest-growing open-source AI skills, ranked by GitHub star growth. One email a week, unsubscribe anytime.
Do this with a working workflow
Pages we wrote and tested that put a skill like this to work — importable n8n templates, outcome guides and Claude Desktop configs.
Weekly Client Report
Every Monday, turn the week’s metrics into a clear client-facing narrative report — what happened, why, what’s next — and email it automatically.
Outcome guideAutomate your weekly client reports
Keep each client’s metrics in a Google Sheet. Every Monday the workflow reads the week’s numbers, writes a plain-English narrative — what moved, why, and what’s next — and emails it to the client.
Outcome guideAutomated marketing performance reports
The workflow reads your performance metrics and writes the narrative for you — this week at a glance, what moved and why, what’s next — then sends it on schedule.
Competitor Watch
Get emailed the moment a rival changes their price or pitch.
Includes the tested workflow and a written setup playbook. Needs: No extra API — just watches the pages you list.
Related skills
More analytics tools founders pair with this one.