Anthropic

Connect Integrations to
Anthropic

Build advanced AI agents with Anthropic. Connect 600+ integrations, automate workflows, and deploy with ease using Metorial.

Back to Anthropic overview

Working with Tools in AI SDK

Understand how Metorial's tools work with the Vercel AI SDK and how to maximize their effectiveness.

What Are Tools?

Tools (also called functions) are capabilities you give to your AI model. With Metorial, these tools represent integrations with external services like Google Calendar, Slack, Notion, and 600+ others.

Accessing Tools

When you create a Metorial session, tools are automatically available via session.tools:

await metorial.withProviderSession(
  metorialAISDK,
  { serverDeployments: ['your-deployment-id'] },
  async (session) => {
    // session.tools contains all tools from your deployment
    console.log(`Available tools: ${session.tools.length}`);
    
    const result = await generateText({
      model: yourModel,
      messages: [...],
      tools: session.tools  // Pass tools to AI SDK
    });
  }
);

How AI SDK Uses Tools

The AI SDK handles the tool execution flow automatically:

  1. Model analyzes request: Determines if tools are needed
  2. Tool selection: Chooses appropriate tools from session.tools
  3. Tool execution: Calls the tools through Metorial
  4. Result integration: Uses tool results to generate response

Controlling Tool Behavior

Use AI SDK's maxSteps parameter to control iterations:

const result = await generateText({
  model: yourModel,
  messages: [...],
  tools: session.tools,
  maxSteps: 5  // Limit to 5 tool call iterations
});

This prevents infinite loops while allowing complex multi-step tasks.

Tool Execution Results

Access detailed information about tool calls:

const result = await generateText({
  model: yourModel,
  messages: [...],
  tools: session.tools
});

// Check what tools were called
console.log('Steps taken:', result.steps);
console.log('Tool calls:', result.toolCalls);
console.log('Tool results:', result.toolResults);

Best Practices

Be specific in prompts: Clear requests help the model choose the right tools

// Good
"Schedule a meeting with John for tomorrow at 2pm"

// Less effective
"Help me with my calendar"

Provide context: Give the model information needed to execute tools effectively

const messages = [
  { role: 'system', content: 'You are a helpful assistant with access to the user\'s calendar and email.' },
  { role: 'user', content: 'Schedule a team sync' }
];

Handle errors gracefully: Some tool calls may fail

const result = await generateText({
  model: yourModel,
  messages: [...],
  tools: session.tools,
  maxSteps: 10
});

if (result.finishReason === 'error') {
  console.error('Tool execution failed');
}

Tool Categories

Metorial provides tools across many categories:

  • Productivity: Google Calendar, Microsoft Outlook, Notion
  • Communication: Slack, Discord, Microsoft Teams
  • Development: GitHub, GitLab, Jira, Linear
  • CRM: Salesforce, HubSpot, Pipedrive
  • And 590+ more integrations

All tools work the same way through the AI SDK interface.

Anthropic on Metorial

Integrate Anthropic Claude with Metorial to give your AI agents access to over 600 powerful integrations through the Model Context Protocol (MCP). Our open-source platform is purpose-built for developers creating intelligent applications with Claude, Anthropic's advanced language model. With Metorial's Python and TypeScript SDKs, connecting Claude to calendars, CRMs, databases, project management tools, and hundreds of other services takes just a couple of lines of code. Stop spending days building custom integrations—Metorial handles authentication, API calls, and data formatting automatically so you can focus on crafting intelligent agent behaviors and user experiences. Whether you're building customer support bots, research assistants, or workflow automation with Claude, Metorial provides the integration infrastructure you need. Our MCP-native approach ensures your Anthropic-powered agents can interact with the tools your users already depend on, creating more valuable and practical AI solutions that integrate seamlessly into existing workflows.

Connect anything. Anywhere.

Ready to build with Metorial?

Let's take your AI-powered applications to the next level, together.

About Metorial

Metorial provides developers with instant access to 600+ MCP servers for building AI agents that can interact with real-world tools and services. Built on MCP, Metorial simplifies agent tool integration by offering pre-configured connections to popular platforms like Google Drive, Slack, GitHub, Notion, and hundreds of other APIs. Our platform supports all major AI agent frameworks—including LangChain, AutoGen, CrewAI, and LangGraph—enabling developers to add tool calling capabilities to their agents in just a few lines of code. By eliminating the need for custom integration code, Metorial helps AI developers move from prototype to production faster while maintaining security and reliability. Whether you're building autonomous research agents, customer service bots, or workflow automation tools, Metorial's MCP server library provides the integrations you need to connect your agents to the real world.

Star us on GitHub