Metorial seamlessly integrates with Vercel's AI SDK, giving your AI applications instant access to over 600 integrations. This guide will walk you through setting up Metorial with the AI SDK.

Prerequisites

• A Metorial account with an API key
• A Vercel AI SDK project
• Node.js installed on your machine

Installation

First, install the required packages:

npm install @metorial/ai-sdk metorial ai

Basic Setup

Here's a minimal example to get started:

import { metorialAiSdk } from '@metorial/ai-sdk';
import { Metorial } from 'metorial';
import { generateText } from 'ai';

const metorial = new Metorial({
  apiKey: process.env.METORIAL_API_KEY
});

await metorial.withProviderSession(
  metorialAiSdk,
  {
    serverDeployments: [process.env.SERVER_DEPLOYMENT_ID]
  },
  async session => {
    const result = await generateText({
      model: yourModel, // your AI SDK model
      messages: [{ role: 'user', content: 'What integrations are available?' }],
      tools: session.tools
    });
    
    console.log(result.text);
  }
);

What's Happening

1. Metorial Client: Creates a connection to Metorial's integration platform
2. Provider Session: Establishes a session with your configured server deployments
3. Tools: Automatically provides 600+ integrations as tools to the AI SDK
4. Tool Calling: The AI SDK handles tool execution through Metorial

Next Steps

• Configure your server deployments in the Metorial dashboard
• Explore available integrations
• Build more complex agentic workflows