Learn how to use Metorial's 600+ integrations as tools in your AI SDK applications.

Basic Tool Usage

When you create a provider session with Metorial, all configured integrations become available as tools:

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

const metorial = new Metorial({
  apiKey: 'your-metorial-api-key'
});

await metorial.withProviderSession(
  metorialAiSdk,
  { serverDeployments: ['your-server-deployment-id'] },
  async session => {
    // session.tools contains all available integrations
    const result = await generateText({
      model: openai('gpt-4'),
      tools: session.tools,
      prompt: 'Create a new issue in GitHub for the bug I just described'
    });
    
    console.log(result.text);
  }
);

Streaming Responses

The AI SDK supports streaming responses with Metorial tools:

import { streamText } from 'ai';

await metorial.withProviderSession(
  metorialAiSdk,
  { serverDeployments: ['your-server-deployment-id'] },
  async session => {
    const result = streamText({
      model: openai('gpt-4'),
      tools: session.tools,
      prompt: 'Summarize my recent emails and create calendar events for follow-ups'
    });
    
    for await (const chunk of result.textStream) {
      process.stdout.write(chunk);
    }
  }
);

Multi-Step Tool Calling

The AI SDK automatically handles multi-step tool calling. Your AI agent can use multiple tools in sequence:

const result = await generateText({
  model: openai('gpt-4'),
  tools: session.tools,
  maxSteps: 10, // Allow up to 10 tool calls
  prompt: 'Check my calendar, find free time this week, and schedule a meeting with the team'
});

What Metorial Handles

Metorial automatically manages:

• Tool authentication and authorization
• API rate limiting
• Error handling and retries
• Tool response formatting
• Session management

You just focus on building your AI application!