How to Use the Braintrust MCP in Vercel AI SDK

Q: How do I connect Braintrust to my Vercel AI SDK project?

Install @ai-sdk/mcp and initialize the MCP client using createMCPClient with your Vinkius HTTP endpoint. Pass the tools directly into streamText or generateText to let your agent call them. Remember to call mcpClient.close() when the session ends.

Q: Do I need to manage model API keys manually in my code?

You do not. Use listenvvars to securely check the AI Gateway configurations that handle your model keys. Vinkius manages the underlying credentials so your frontend never exposes them.

Pipe Braintrust prompt variables and test datasets directly into your live Vercel AI SDK frontend via this MCP integration.

See Vinkius in Action

Works with every AI agent you already use

…and any MCP-compatible client

MCP Servers - Free for Subscribers

Connect Braintrust MCP to Vercel AI SDK

Create your Vinkius account to connect Braintrust to Vercel AI 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 Braintrust with Vercel AI SDK

Ask AI about this MCP

ChatGPT

Claude

Perplexity

Live-stream prompt updates with Vercel AI SDK

The Braintrust MCP Server lets your application fetch raw text templates and variable contexts using `get_prompt` or `list_prompts` during a live user session. Stop hardcoding system instructions in your Next.js codebase. Your users get the latest model instructions instantly. Because this integration runs on the edge, you avoid cold starts and keep your latency low.

Feed live test cases into evaluations

This Braintrust MCP Server lets you append new test cases into your evaluation matrix using the `insert_dataset_row` tool. When users flag bad outputs in your UI, you need to capture those examples immediately. This setup turns real-world edge cases into test data. You can quickly run automated tests against the new rows without leaving your terminal.

Track experiments during local runs

Running `create_experiment` directly from your development server allows you to spin up new test cycles on the fly. You can watch your model scores update inside your terminal as you tweak your system prompts. This helps you spot regressions before pushing code. Your team gets instant visibility into performance changes.

Setup guide

Set up Braintrust MCP in Vercel AI SDK

Prerequisites

Node.js 18+ and a TypeScript project
ai + @modelcontextprotocol/sdk packages
Active Vinkius subscription with a valid endpoint token

1

Install dependencies

Run npm install ai @modelcontextprotocol/sdk plus your preferred model provider (e.g. @ai-sdk/openai).
2

Create the Streamable HTTP transport

Use StreamableHTTPClientTransport with your Vinkius endpoint URL. Replace [YOUR_TOKEN_HERE] with your token from cloud.vinkius.com.
3

Discover and use tools

Call mcpClient.tools() to auto-discover all Braintrust tools. Pass them directly to generateText() or streamText() — no manual schema definitions needed.
4

Works with any model provider

Swap openai("gpt-4o") for any AI SDK provider — Anthropic, Google, Mistral. The MCP tools work identically across all supported models.

index.ts

import { experimental_createMCPClient as createMCPClient } from "ai";
import { StreamableHTTPClientTransport } from "@modelcontextprotocol/sdk/client/streamableHttp";
import { generateText } from "ai";
import { openai } from "@ai-sdk/openai";

const transport = new StreamableHTTPClientTransport(
  new URL("https://edge.vinkius.com/[YOUR_TOKEN_HERE]/mcp")
);

const mcpClient = await createMCPClient({ transport });
const tools = await mcpClient.tools();

const { text } = await generateText({
  model: openai("gpt-4o"),
  tools,
  prompt: "List recent Braintrust transactions",
});

console.log(text);
await mcpClient.close();

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 Braintrust. 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 Braintrust 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 Braintrust MCP in Vercel AI SDK

Install `@ai-sdk/mcp` and initialize the MCP client using `createMCPClient` with your Vinkius HTTP endpoint. Pass the tools directly into `streamText` or `generateText` to let your agent call them. Remember to call `mcpClient.close()` when the session ends.

Yes, the client is fully compatible with Edge runtimes. It uses standard HTTP transports to fetch prompts via `get_prompt` without heavy dependencies. This keeps your serverless functions fast and lightweight.

The SDK treats the server tools as standard functions. When your agent decides it needs to fetch a dataset using `get_dataset`, the SDK executes the tool and streams the JSON payload directly back to the model context.

You do not. Use `list_env_vars` to securely check the AI Gateway configurations that handle your model keys. Vinkius manages the underlying credentials so your frontend never exposes them.

Your prompt templates and dataset rows are protected inside a secure V8 Isolate sandbox. Vinkius hosts the endpoint as an ephemeral process, meaning no evaluation data is stored on our servers after the execution finishes. No keys or dataset schemas are ever persisted.

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