How to Use the NotCo MCP in Vercel AI SDK

Q: How do I handle streaming errors when connecting NotCo to Vercel AI SDK?

If Giuseppe AI encounters an error during createformulation, the MCP Server passes the error block directly through the stream. You should catch this in your streamText call and render a fallback UI so food scientists know if an ingredient molecular profile is missing.

Q: How does Vercel AI SDK manage NotCo tool calling for complex recipes?

You pass the tools from mcpClient.tools() directly into the tools parameter of generateText. Your agent will automatically decide whether to call searchflavormatches or analyzenutrition based on what the food scientist asks for in the chat.

Q: What is the best way to close the connection after running a simulation?

Always invoke mcpClient.close() inside your clean-up blocks or at the end of your serverless execution. Leaving the connection open wastes resources and might cause subsequent Giuseppe AI formulation requests to hang.

Build live plant-based food R&D apps using Vercel AI SDK and this NotCo MCP Server.

See Vinkius in Action

Works with every AI agent you already use

…and any MCP-compatible client

MCP Servers - Free for Subscribers

Connect NotCo MCP to Vercel AI SDK

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

Ask AI about this MCP

ChatGPT

Claude

Perplexity

Stream live molecular matching to your frontend

This MCP Server exposes `search_flavor_matches` to let your Vercel AI SDK app stream real-time plant-based alternatives directly to your user's UI. Instead of making users stare at a blank screen while Giuseppe AI processes complex chemical compounds, the matching ingredients appear on screen piece by piece. You can combine this with `get_ingredient` to fetch deep molecular profiles of plant substitutes instantly. The Edge-compatible setup ensures your food science dashboard stays fast, rendering raw data without blocking the main thread or requiring heavy server-side middleware.

Real-time sensory simulation rendering

This MCP Server uses `run_sensory_test` to simulate how consumers will react to your plant-based prototype, streaming the sensory scores straight to your React components. Your developers can build interactive sliders where food scientists tweak ingredient ratios and watch the simulated texture and taste graphs update live. By hook-linking this to `list_sensory_profiles`, your Vercel AI SDK application lets R&D teams compare their plant-based prototypes against dairy or meat benchmarks side-by-side. The UI updates instantly as Giuseppe AI evaluates the chemical stability of the mixture.

Instant cost and nutrition calculation

This MCP Server provides `estimate_cost` to predict the commercial production price of your plant-based recipes on the fly. When a user modifies an ingredient in your AI SDK frontend, the cost projection and `analyze_nutrition` metrics update dynamically. This prevents food developers from designing perfect plant-based cheeses that are too expensive to manufacture. By feeding these tools directly into your streaming interface, your team gets immediate financial and nutritional feedback before saving the recipe.

Setup guide

Set up NotCo 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 NotCo 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 NotCo 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 NotCo. 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 NotCo 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 NotCo MCP in Vercel AI SDK

If Giuseppe AI encounters an error during `create_formulation`, the MCP Server passes the error block directly through the stream. You should catch this in your `streamText` call and render a fallback UI so food scientists know if an ingredient molecular profile is missing.

Yes. The server runs on Vinkius infrastructure, meaning your Edge Functions only need to make lightweight HTTP requests to query `run_sensory_test` or `list_ingredients`. This keeps your Next.js startup times under 50ms.

You pass the tools from `mcpClient.tools()` directly into the `tools` parameter of `generateText`. Your agent will automatically decide whether to call `search_flavor_matches` or `analyze_nutrition` based on what the food scientist asks for in the chat.

Always invoke `mcpClient.close()` inside your clean-up blocks or at the end of your serverless execution. Leaving the connection open wastes resources and might cause subsequent Giuseppe AI formulation requests to hang.

Vinkius hosts this server in an isolated V8 sandbox, ensuring your custom recipes and ingredient lists never leak. All data sent to `create_formulation` is encrypted in transit and processed ephemerally, meaning your proprietary food IP remains entirely yours.

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