How to Use the Extracta MCP in Vercel AI SDK

Q: How do I handle file uploads using Extracta with the Vercel AI SDK?

You pass the target file URL to uploadfileurl directly from your serverless function. Your Vercel AI SDK agent then polls getresults using the MCP server to stream the parsed JSON structure back to your frontend.

Q: Can I update active Extracta extraction rules through the Vercel AI SDK?

Yes, your agent uses updateextraction to modify schema requirements on the fly. This updates the target extraction process without requiring a restart of your Vercel deployment.

Q: What is the best way to run Extracta categorization in Vercel AI SDK?

Define your target categories via createclassification within your serverless route. The SDK then reads the document type using getclassificationresults to determine the next UI state.

Q: How do I clean up completed extraction processes?

Call deleteextraction when your user finishes their session. This invalidates the extraction ID and prevents further file uploads to that specific configuration on the MCP server.

Q: Where do my uploaded documents go during Extracta processing?

Files sent via uploadfileurl are processed in ephemeral V8 sandboxes managed by Vinkius. Extracta deletes the document data from active memory once getresults delivers the final structured JSON payload to your SDK client.

Get structured JSON from raw documents inside your Vercel AI SDK chat interface using this 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 Extracta MCP to Vercel AI SDK

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

Ask AI about this MCP

ChatGPT

Claude

Perplexity

Stream document data live with Vercel AI SDK

The `upload_file_url` tool sends files directly from your React frontend to Extracta for immediate processing. Your AI client initiates the upload and monitors the document status without blocking the main UI thread. By calling `get_results` inside a streaming text loop, you feed extracted fields straight to the user's screen. They watch invoice totals and line items render item-by-item instead of staring at a blank screen.

Modify extraction rules during active Edge sessions

The `update_extraction` tool lets your agent modify data schemas mid-session based on user feedback. If a user points out a missing field in a contract, the SDK triggers an update to the extraction rules without rebuilding your Next.js application. Your Vercel AI SDK application immediately applies these new parameters to subsequent runs. This means you do not need to redeploy your edge function just to capture a new metadata field.

Route documents dynamically inside the user interface

The `create_classification` tool maps incoming PDFs to your predefined categories during a live chat session. When a user drops a file into the chat, this MCP tool routes the file path to verify if it is an invoice or a tax form. Running `get_classification_results` lets your Vercel AI SDK interface render custom UI components based on the document type. An invoice displays a payment button, while a contract renders a signature block instantly.

Setup guide

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

You pass the target file URL to `upload_file_url` directly from your serverless function. Your Vercel AI SDK agent then polls `get_results` using the MCP server to stream the parsed JSON structure back to your frontend.

Yes, your agent uses `update_extraction` to modify schema requirements on the fly. This updates the target extraction process without requiring a restart of your Vercel deployment.

Define your target categories via `create_classification` within your serverless route. The SDK then reads the document type using `get_classification_results` to determine the next UI state.

Call `delete_extraction` when your user finishes their session. This invalidates the extraction ID and prevents further file uploads to that specific configuration on the MCP server.

Files sent via `upload_file_url` are processed in ephemeral V8 sandboxes managed by Vinkius. Extracta deletes the document data from active memory once `get_results` delivers the final structured JSON payload to your SDK client.

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