Langfuse integration
Open source LLM engineering platform. Traces, evals, prompt management and metrics to debug and improve your LLM application.
What your agents can do with Langfuse
4 pre-built Langfuse tool steps, ready to drop into any agent. Anything they do not cover, an agent can reach with an authenticated API call.
Actions (2)
Log Trace
Log a new trace in LangFuse with details.
Add Feedback
Attach user feedback to an existing trace in Langfuse.
Triggers (2)
New Trace Received
Emit new event when a new trace is recorded in Langfuse.
New Score Received
Emit new event when user feedback (score) is submitted on a trace in Langfuse.
Set up Langfuse in Relevance AI
Do this once. Every agent in your workspace can then use the connection.
Create a Relevance AI account
Sign in, or create a free account. You do not need to be a developer to set this up.
Connect your Langfuse account
In Relevance AI, open Integrations and add Langfuse, then paste your Region, Public Key, and Secret Key. You will find it in your Langfuse account settings.
Use Langfuse in an agent
Create a tool, add a Langfuse tool step, and connect it to an agent. The agent fills in the step's inputs each time it runs.
Put Langfuse to work in your agents
Connect Langfuse alongside 1,800+ other integrations and let your agents run the busywork.