How to Use the Langfuse (LLM Tracing & Evals) MCP in OpenAI Agents SDK

Q: How do I connect Langfuse (LLM Tracing & Evals) to the OpenAI Agents SDK?

Install the package using pip install openai-agents. Create an MCPServerStreamableHttp instance with your Vinkius URL and pass it to the mcpservers list in your Agent constructor. Set cacheToolsList=True to speed up tool discovery for the MCP Server.

Q: Can the OpenAI Agents SDK read my Langfuse (LLM Tracing & Evals) prompt templates?

Yes. The agent uses the listprompts tool to fetch active templates. This keeps your Python code clean and your prompts managed centrally.

Q: Does Langfuse (LLM Tracing & Evals) work with OpenAI Agents SDK guardrails?

Yes, it does. Your guardrail functions can call gettrace to inspect previous turns before approving the next action. This prevents unsafe handoffs between specialized agents.

Q: How do I log a user rating from my agent interface?

Trigger the createscore tool. The agent passes the integer rating and maps it directly to the current trace ID.

Connect the Langfuse (LLM Tracing & Evals) MCP Server to your OpenAI Agents SDK environment to monitor traces and audit scores in production.

See Vinkius in Action

Works with every AI agent you already use

…and any MCP-compatible client

MCP Servers - Free for Subscribers

Connect Langfuse (LLM Tracing & Evals) MCP to OpenAI Agents SDK

Create your Vinkius account to connect Langfuse (LLM Tracing & Evals) to OpenAI Agents SDK and route execution through our secure gateway. The platform manages server hosting, runtime updates, and security layers. Configuration requires no manual server provisioning.

GDPR Free for Subscribers

Setup Langfuse (LLM Tracing & Evals) with OpenAI Agents SDK

Ask AI about this MCP

ChatGPT

Claude

Perplexity

Track LLM Traces with OpenAI Agents SDK

`list_traces` and `get_trace` pull complete telemetry graphs directly into your agent runtime. You get full visibility into what your multi-agent system is doing behind the scenes. Every step, prompt, and output gets logged via this MCP integration. Handoffs between specialized agents generate complex execution paths. Using these tools lets your guardrail mechanisms inspect past interactions before proceeding. The system catches unsafe patterns by analyzing historical trace data instead of guessing.

Audit Evaluation Scores

`create_score` attaches human feedback or automated pipeline metrics to a specific trace or observation. Your agent can read these metrics later using `list_scores`. This loop creates a strict quality baseline for your deployed product. Production systems require rigid safety constraints. When an agent flags a bad generation, it logs the failure instantly. You define the exact parameters for acceptable outputs.

Monitor Daily Usage and Prompts

`get_daily_metrics` generates rolled-up USD cost and aggregated latency statistics across your active sessions. Cost control matters when autonomous systems run continuously. The agent checks these figures to ensure it stays under budget limits. Managing templates happens through `list_prompts`. You extract actively managed prompt versions without hardcoding them into your Python scripts. Updating a prompt in the dashboard immediately reflects in the agent's next execution cycle.

Setup guide

Set up Langfuse (LLM Tracing & Evals) MCP in OpenAI Agents SDK

Prerequisites

Python 3.10+ installed
openai-agents package (pip install openai-agents)
Active Vinkius subscription with a valid endpoint token

1

Install the SDK

Run pip install openai-agents to install the OpenAI Agents SDK. The MCP integration is built-in — no extra dependencies needed.
2

Connect via SSE transport

Use MCPServerSse with your Vinkius endpoint URL. Replace [YOUR_TOKEN_HERE] with your token from cloud.vinkius.com. The SDK auto-discovers all Langfuse (LLM Tracing & Evals) tools at runtime.
3

Create your Agent

Pass the MCP to Agent(mcp_servers=[server]). The agent receives Langfuse (LLM Tracing & Evals) tools as native definitions — JSON schemas resolve automatically.
4

Run the agent

Call Runner.run(agent, prompt) to execute. The agent invokes the appropriate Langfuse (LLM Tracing & Evals) tools and returns structured results. Copy the full example on the right to get started.

agent.py

import asyncio
from agents import Agent, Runner
from agents.mcp import MCPServerSse

async def main():
    async with MCPServerSse(
        url="https://edge.vinkius.com/[YOUR_TOKEN_HERE]/mcp"
    ) as server:
        agent = Agent(
            name="Langfuse (LLM Tracing & Evals) Agent",
            instructions="You have access to Langfuse (LLM Tracing & Evals) tools.",
            mcp_servers=[server],
        )
        result = await Runner.run(agent, "List recent transactions")
        print(result.final_output)

asyncio.run(main())

Get your connection token →

Independent Platform Disclaimer: Vinkius is an independent platform and is not affiliated with, endorsed by, sponsored by, verified by, or otherwise authorized by Langfuse. All third-party trademarks, logos, and brand names are the property of their respective owners. Their use on this website is strictly for informational purposes to identify service compatibility and interoperability.

Why Choose Vinkius

Vinkius connects your tools to AI with real-time monitoring and automatic cost savings — all from one dashboard.

Connect Langfuse (LLM Tracing & Evals) now

Real-time monitoring

Live

visibility into every interaction

Connect your favorite tools to your AI and see exactly what's happening — every request, every response, in real time.

Built-in savings

60%

lower AI costs

Vinkius compresses data between your apps and your AI automatically. Lower bills every month — no configuration required.

Single dashboard

One

place for every integration

Every tool your AI connects to, managed from a single screen. One account, complete control.

Common questions about Langfuse (LLM Tracing & Evals) MCP in OpenAI Agents SDK

Install the package using `pip install openai-agents`. Create an `MCPServerStreamableHttp` instance with your Vinkius URL and pass it to the `mcp_servers` list in your Agent constructor. Set `cacheToolsList=True` to speed up tool discovery for the MCP Server.

Yes. The agent uses the `list_prompts` tool to fetch active templates. This keeps your Python code clean and your prompts managed centrally.

Yes, it does. Your guardrail functions can call `get_trace` to inspect previous turns before approving the next action. This prevents unsafe handoffs between specialized agents.

Trigger the `create_score` tool. The agent passes the integer rating and maps it directly to the current trace ID.

Vinkius runs this MCP Server in a zero-trust V8 Isolate Sandbox. Your telemetry graphs, span contexts, and prompt histories remain ephemeral during transit. Authentication requires only your single endpoint token.

Use it with your favorite AI tools

Connect this server to Cursor, Claude, VS Code, and more.

OpenAI Agents SDK sdk-python

Google ADK sdk-python

Pydantic AI sdk-python

Vercel AI SDK sdk-typescript