How to Use the Langfuse (LLM Tracing & Evals) MCP in Cline

Q: Can Cline pull live traces from Langfuse (LLM Tracing & Evals) to debug my code?

Yes, Cline calls gettrace to fetch the complete telemetry and nested graph for any production run. The agent reads this context to locate bugs in your LLM pipeline and write patches.

Q: How does Cline write evaluation scores back to Langfuse (LLM Tracing & Evals)?

Cline uses the createscore tool to attach automated evaluation metrics or human feedback directly to a specific trace. This makes it easy to log test results from your local VS Code environment.

Q: Can Cline analyze user sessions using Langfuse (LLM Tracing & Evals)?

Yes, Cline invokes listsessions to pull high-level user session entities. It then digs into individual traces to diagnose long-running conversational issues.

Let Cline pull live LLM traces and evaluation scores to rewrite prompts and patch production bugs directly in VS Code.

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 Cline

Create your Vinkius account to connect Langfuse (LLM Tracing & Evals) to Cline 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 Cline

Ask AI about this MCP

ChatGPT

Claude

Perplexity

Automate prompt fixes using the Langfuse MCP Server

To fix broken LLM outputs, Cline uses `list_prompts` to inspect your active prompt templates and find mismatches with your application code. The agent compares production templates against local source files to spot drifted variables or deprecated parameters. If a prompt is underperforming, the agent pulls down raw generation details using `get_observation` to see what the model actually output. Cline then rewrites the prompt, runs local tests, and prepares the commit without manual intervention.

Analyze user sessions and traces recursively

Debugging multi-turn conversational agents requires deep context, so Cline executes this MCP tool `list_sessions` to group related LLM calls together. This lets the agent inspect the entire history of a single user interaction to understand where a conversation went off the rails. For granular analysis, the agent calls `get_trace` to retrieve the nested execution graph of any specific session step. Cline handles the heavy lifting of parsing the JSON payload to find the exact node that failed.

Track human feedback and evaluation scores

To audit the quality of your LLM outputs, the agent uses `list_scores` to retrieve evaluation metrics and human feedback stars. Cline analyzes these scores to identify which system prompts are generating poor user experiences. You can also have the agent run `create_score` to programmatically log evaluation results back to Langfuse during local test runs via this MCP connection. It keeps your evaluation database updated without manual data entry.

Setup guide

Set up Langfuse (LLM Tracing & Evals) MCP in Cline

Prerequisites

VS Code with Cline extension installed
Active Vinkius subscription with a valid endpoint token

1

Open Cline MCP settings

Click the Cline icon in the VS Code sidebar to open the Cline panel. Then click the MCP Servers icon (server stack) at the top-right corner of the panel.
2

Add a remote server

Click "Remote Servers" at the top, then click "Add Remote MCP". In the Name field, type langfuse-llm-tracing-evals-mcp. In the URL field, paste your Vinkius endpoint: https://edge.vinkius.com/[YOUR_TOKEN_HERE]/mcp. Get your token from cloud.vinkius.com.
3

Enable the server

After saving, the server appears in the Cline MCP panel. Toggle the switch to enable it. The status indicator turns green when the connection is live.
4

Start using tools

Return to the Cline chat and ask: "Check my latest Langfuse (LLM Tracing & Evals) refund status." Cline will discover the available tools and request your approval before invoking each one — giving you full control over every action.

Cline MCP Settings

{
  "mcpServers": {
    "langfuse-llm-tracing-evals-mcp": {
      "url": "https://edge.vinkius.com/[YOUR_TOKEN_HERE]/mcp"
    }
  }
}

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 Cline

Yes, Cline calls `get_trace` to fetch the complete telemetry and nested graph for any production run. The agent reads this context to locate bugs in your LLM pipeline and write patches.

Cline uses the `create_score` tool to attach automated evaluation metrics or human feedback directly to a specific trace. This makes it easy to log test results from your local VS Code environment.

Click the MCP icon in the Cline sidebar and search for the server in the marketplace tab for a one-click install. You can also manually add the configuration block to your `cline_mcp_settings.json` file.

Yes, Cline invokes `list_sessions` to pull high-level user session entities. It then digs into individual traces to diagnose long-running conversational issues.

Your LLM traces, observations, scores, and prompt templates flow directly through a secure, ephemeral V8 isolate sandbox. Vinkius runs the MCP server in an isolated environment and does not store your API keys or data payloads, ensuring your operational telemetry remains private.

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