How to integrate Tave MCP with Vercel AI SDK v6

GET STARTED FOR FREE
Framework Integration Gradient
Tave Logo
Vercel AI SDK Logo
divider

Introduction

This guide walks you through connecting Tave to Vercel AI SDK v6 using the Composio tool router. By the end, you'll have a working Tave agent that can list all open jobs assigned to me, add a new client to tave contacts, update status of a job to 'completed' through natural language commands.

This guide will help you understand how to give your Vercel AI SDK agent real control over a Tave account through Composio's Tave MCP server.

Before we dive in, let's take a quick look at the key ideas and tools involved.

TL;DR

Here's what you'll learn:
  • How to set up and configure a Vercel AI SDK agent with Tave integration
  • Using Composio's Tool Router to dynamically load and access Tave tools
  • Creating an MCP client connection using HTTP transport
  • Building an interactive CLI chat interface with conversation history management
  • Handling tool calls and results within the Vercel AI SDK framework

What is Vercel AI SDK?

The Vercel AI SDK is a TypeScript library for building AI-powered applications. It provides tools for creating agents that can use external services and maintain conversation state.

Key features include:

  • streamText: Core function for streaming responses with real-time tool support
  • MCP Client: Built-in support for Model Context Protocol via @ai-sdk/mcp
  • Step Counting: Control multi-step tool execution with stopWhen: stepCountIs()
  • OpenAI Provider: Native integration with OpenAI models

What is the Tave MCP server, and what's possible with it?

The Tave MCP server is an implementation of the Model Context Protocol that connects your AI agent and assistants like Claude, Cursor, etc directly to your Tave account. It provides structured and secure access so your agent can perform Tave operations on your behalf.

Supported Tools & Triggers

Tools
Apply Payment to OrderTool to apply a payment to a specific order in Tave.
Create BrandTool to create a new brand in VSCO Workspace.
Create Custom FieldTool to create a custom field in Tave for Jobs or Contacts.
Create Discount TypeTool to create a new discount type in Tave.
Create GalleryTool to create a new gallery in Tave (VSCO Workspace).
Create Address Book EntryTool to create an entry in the address book.
Create DiscountTool to create a discount in Tave.
Create EventTool to create a new event in Tave (VSCO Workspace).
Create Event TypeTool to create a new event type in Tave.
Create FileTool to create a file in Tave.
Create JobTool to create a new job in Tave (VSCO Workspace).
Create Job Closed ReasonTool to create a job closed reason in Tave.
Create Job ContactTool to create a job contact in Tave.
Create Job OrderTool to create an order on a job in Tave (VSCO Workspace).
Create Job RoleTool to create a job role in Tave (VSCO Workspace).
Create Job TypeTool to create a new job type in Tave.
Create Job WorksheetTool to create a job using a worksheet in Tave (VSCO Workspace).
Create Lead SourceTool to create a lead source in Tave.
Create Lead StatusTool to create a lead status in Tave.
Create NoteTool to create a note in Tave (VSCO Workspace).
Create PaymentTool to create a payment in Tave.
Create ProductTool to create a product in Tave.
Create Product TypeTool to create a product type in Tave.
Create Profit CenterTool to create a profit center in Tave.
Create RestHook SubscriptionTool to create a RestHook webhook subscription in Tave.
Create Tax RateTool to create a tax rate in Tave.
Create UserTool to create a new user in Tave.
Delete Address Book EntryTool to delete an entry in the address book.
Delete BrandTool to delete a brand by its ULID.
Delete Custom FieldTool to delete a custom field by its ULID identifier.
Delete Event TypeTool to delete an event type by its ULID identifier.
Delete GalleryTool to permanently delete a gallery by its ULID.
Delete DiscountTool to delete a specific discount by its ULID.
Delete Discount TypeTool to delete a specific discount type by its ULID.
Delete EventTool to delete an event by its ULID identifier.
Delete JobTool to delete a job by its ULID identifier.
Delete Job Closed ReasonTool to delete a job closed reason by its ULID identifier.
Delete Job ContactTool to delete a job contact by its ULID.
Delete Job RoleTool to delete a job role by its ULID identifier.
Delete Job TypeTool to delete a job type by its ULID.
Delete Lead SourceTool to delete a specific lead source by its ULID.
Delete Lead StatusTool to delete a specific lead status by its ULID.
Delete NoteTool to delete a note by its ULID identifier.
Delete PaymentTool to delete a specific payment by its ULID.
Delete ProductTool to delete a specific product by its ULID.
Delete Product TypeTool to delete a specific product type by its ULID.
Delete Profit CenterTool to delete a profit center by its ULID.
Delete RestHookTool to unsubscribe from and delete a RestHook webhook subscription by its ULID.
Delete Tax GroupTool to delete a specific tax group by its ULID.
Delete Tax RateTool to delete a specific tax rate by its ULID.
Delete UserTool to delete a user by its ULID.
Get BrandTool to retrieve a specific brand from Tave by its ULID.
Get EventTool to get a specific event by its ID.
Get Event TypeTool to retrieve a specific event type by ID.
Get GalleryTool to retrieve a specific gallery by ID from Tave/VSCO Workspace.
Get Address Book EntityTool to fetch a specific Address Book entity by its ULID.
Get Address Book AddressTool to fetch the address associated with a given location ID.
Get Custom FieldTool to fetch a specific custom field by ID from Tave.
Get DiscountTool to retrieve a specific discount by its ID.
Get Discount TypeTool to get a specific discount type by ID.
Get JobTool to fetch a specific job by ID from Tave.
Get Job Closed ReasonTool to retrieve a specific job closed reason by its ID.
Get Job ContactTool to fetch a specific job contact by ID from Tave.
Get Job RoleTool to retrieve a specific job role by its ID.
Get Job TypeTool to retrieve a specific job type by ID.
Get Job WorksheetTool to fetch a Job Worksheet for a specific job by job ID.
Get Lead SourceTool to retrieve a specific lead source by its ID.
Get Lead StatusTool to retrieve a specific lead status by its ID.
Get My StudioTool to retrieve information about your studio from Tave.
Get NoteTool to fetch a specific note by ID from Tave.
Get OrderTool to fetch a specific order by ID from Tave.
Get PaymentTool to fetch a specific payment by ID from Tave.
Get Payment MethodTool to retrieve a specific payment method by its ID.
Get ProductTool to retrieve a specific product by its ID.
Get Product TypeTool to retrieve a specific product type by its ID.
Get Profit CenterTool to retrieve a specific profit center by its ID.
Get Tax RateTool to retrieve a specific tax rate by its ULID.
Get TimezoneTool to retrieve a specific timezone by its ULID from Tave.
Get User by IDTool to retrieve a specific user by their ULID identifier.
List Address Book ContactsTool to list contacts in the studio's address book.
List BrandsTool to list brands of a studio.
List Discount TypesTool to get a list of discount types from Tave.
List Studio EventsTool to list a studio's events from Tave.
List Event TypesTool to list a studio's event types.
List FilesTool to list files in Tave.
List Resource Custom FieldsTool to list custom fields configured on a studio.
List Resource DiscountTool to get a list of discounts from Tave.
List GalleriesTool to list all galleries for a studio.
List Resource JobTool to list jobs of a studio.
List Job Closed ReasonsTool to list a studio's job closed reasons.
List Resource Job ContactTool to list a job's contacts.
List Resource Job OrdersTool to list orders on a given job.
List Job PaymentsTool to get a list of payments for a specific job.
List Job RolesTool to list a studio's job roles.
List Resource Job TypeTool to list job types of a studio.
List Resource Lead SourceTool to list lead sources of a studio.
List Resource Lead StatusTool to list lead statuses of a studio.
List Resource NoteTool to list notes from Tave.
List Resource OrderTool to get a list of all orders.
List Resource PaymentTool to get a list of payments from Tave.
List Resource Payment MethodTool to get a list of payment methods from Tave.
List Resource ProductTool to get a list of products from Tave.
List Product TypesTool to get a list of product types from Tave.
List Resource Profit CenterTool to list profit centers of a studio.
List REST HooksTool to list REST hooks configured in Tave.
List Resource Tax GroupsTool to get a list of Tax Groups from Tave.
List Resource Tax RatesTool to get a list of tax rates from Tave.
List Resource TimezoneTool to get a list of timezones from Tave.
List Resource UserTool to list users of a studio.
Update Address Book EntryTool to update a specific entry in the Tave address book.
Update BrandTool to update a specific brand in Tave.
Update Event TypeTool to update an existing event type in Tave.
Update Custom FieldTool to update a specific custom field in Tave.
Update EventTool to update a specific event in Tave.
Update GalleryTool to update a specific gallery in Tave (VSCO Workspace).
Update JobTool to update a specific job in Tave (VSCO Workspace).
Update Job Closed ReasonTool to update a job closed reason in Tave.
Update Job ContactTool to update a specific job contact in Tave.
Update Job RoleTool to update a job role in Tave.
Update Job TypeTool to update an existing job type in Tave.
Update Lead SourceTool to update a specific lead source in Tave.
Update Lead StatusTool to update a specific lead status in Tave.
Update NoteTool to update a specific note in Tave.
Update OrderTool to update a specific order in Tave (VSCO Workspace).
Update Profit CenterTool to update a specific profit center in Tave.
Update Tax GroupTool to update a specific tax group in Tave.
Update Tax RateTool to update a specific tax rate in Tave.
Update UserTool to update a specific user in Tave.
Update Studio Readonly ModeTool to set studio readonly mode in Tave.

What is the Composio tool router, and how does it fit here?

What is Tool Router?

Composio's Tool Router helps agents find the right tools for a task at runtime. You can plug in multiple toolkits (like Gmail, HubSpot, and GitHub), and the agent will identify the relevant app and action to complete multi-step workflows. This can reduce token usage and improve the reliability of tool calls. Read more here: Getting started with Tool Router

The tool router generates a secure MCP URL that your agents can access to perform actions.

How the Tool Router works

The Tool Router follows a three-phase workflow:

  1. Discovery: Searches for tools matching your task and returns relevant toolkits with their details.
  2. Authentication: Checks for active connections. If missing, creates an auth config and returns a connection URL via Auth Link.
  3. Execution: Executes the action using the authenticated connection.

Step-by-step Guide

Prerequisites

Before you begin, make sure you have:
  • Node.js and npm installed
  • A Composio account with API key
  • An OpenAI API key

Getting API Keys for OpenAI and Composio

OpenAI API Key
  • Go to the OpenAI dashboard and create an API key. You'll need credits to use the models, or you can connect to another model provider.
  • Keep the API key safe.
Composio API Key
  • Log in to the Composio dashboard.
  • Navigate to your API settings and generate a new API key.
  • Store this key securely as you'll need it for authentication.

Install required dependencies

bash
npm install @ai-sdk/openai @ai-sdk/mcp @composio/core ai dotenv

First, install the necessary packages for your project.

What you're installing:

  • @ai-sdk/openai: Vercel AI SDK's OpenAI provider
  • @ai-sdk/mcp: MCP client for Vercel AI SDK
  • @composio/core: Composio SDK for tool integration
  • ai: Core Vercel AI SDK
  • dotenv: Environment variable management

Set up environment variables

bash
OPENAI_API_KEY=your_openai_api_key_here
COMPOSIO_API_KEY=your_composio_api_key_here
COMPOSIO_USER_ID=your_user_id_here

Create a .env file in your project root.

What's needed:

  • OPENAI_API_KEY: Your OpenAI API key for GPT model access
  • COMPOSIO_API_KEY: Your Composio API key for tool access
  • COMPOSIO_USER_ID: A unique identifier for the user session

Import required modules and validate environment

typescript
import "dotenv/config";
import { openai } from "@ai-sdk/openai";
import { Composio } from "@composio/core";
import * as readline from "readline";
import { streamText, type ModelMessage, stepCountIs } from "ai";
import { createMCPClient } from "@ai-sdk/mcp";

const composioAPIKey = process.env.COMPOSIO_API_KEY;
const composioUserID = process.env.COMPOSIO_USER_ID;

if (!process.env.OPENAI_API_KEY) throw new Error("OPENAI_API_KEY is not set");
if (!composioAPIKey) throw new Error("COMPOSIO_API_KEY is not set");
if (!composioUserID) throw new Error("COMPOSIO_USER_ID is not set");

const composio = new Composio({
  apiKey: composioAPIKey,
});
What's happening:
  • We're importing all necessary libraries including Vercel AI SDK's OpenAI provider and Composio
  • The dotenv/config import automatically loads environment variables
  • The MCP client import enables connection to Composio's tool server

Create Tool Router session and initialize MCP client

typescript
async function main() {
  // Create a tool router session for the user
  const session = await composio.create(composioUserID!, {
    toolkits: ["tave"],
  });

  const mcpUrl = session.mcp.url;
What's happening:
  • We're creating a Tool Router session that gives your agent access to Tave tools
  • The create method takes the user ID and specifies which toolkits should be available
  • The returned mcp object contains the URL and authentication headers needed to connect to the MCP server
  • This session provides access to all Tave-related tools through the MCP protocol

Connect to MCP server and retrieve tools

typescript
const mcpClient = await createMCPClient({
  transport: {
    type: "http",
    url: mcpUrl,
    headers: session.mcp.headers, // Authentication headers for the Composio MCP server
  },
});

const tools = await mcpClient.tools();
What's happening:
  • We're creating an MCP client that connects to our Composio Tool Router session via HTTP
  • The mcp.url provides the endpoint, and mcp.headers contains authentication credentials
  • The type: "http" is important - Composio requires HTTP transport
  • tools() retrieves all available Tave tools that the agent can use

Initialize conversation and CLI interface

typescript
let messages: ModelMessage[] = [];

console.log("Chat started! Type 'exit' or 'quit' to end the conversation.\n");
console.log(
  "Ask any questions related to tave, like summarize my last 5 emails, send an email, etc... :)))\n",
);

const rl = readline.createInterface({
  input: process.stdin,
  output: process.stdout,
  prompt: "> ",
});

rl.prompt();
What's happening:
  • We initialize an empty messages array to maintain conversation history
  • A readline interface is created to accept user input from the command line
  • Instructions are displayed to guide the user on how to interact with the agent

Handle user input and stream responses with real-time tool feedback

typescript
rl.on("line", async (userInput: string) => {
  const trimmedInput = userInput.trim();

  if (["exit", "quit", "bye"].includes(trimmedInput.toLowerCase())) {
    console.log("\nGoodbye!");
    rl.close();
    process.exit(0);
  }

  if (!trimmedInput) {
    rl.prompt();
    return;
  }

  messages.push({ role: "user", content: trimmedInput });
  console.log("\nAgent is thinking...\n");

  try {
    const stream = streamText({
      model: openai("gpt-5"),
      messages,
      tools,
      toolChoice: "auto",
      stopWhen: stepCountIs(10),
      onStepFinish: (step) => {
        for (const toolCall of step.toolCalls) {
          console.log(`[Using tool: ${toolCall.toolName}]`);
          }
          if (step.toolCalls.length > 0) {
            console.log(""); // Add space after tool calls
          }
        },
      });

      for await (const chunk of stream.textStream) {
        process.stdout.write(chunk);
      }

      console.log("\n\n---\n");

      // Get final result for message history
      const response = await stream.response;
      if (response?.messages?.length) {
        messages.push(...response.messages);
      }
    } catch (error) {
      console.error("\nAn error occurred while talking to the agent:");
      console.error(error);
      console.log(
        "\nYou can try again or restart the app if it keeps happening.\n",
      );
    } finally {
      rl.prompt();
    }
  });

  rl.on("close", async () => {
    await mcpClient.close();
    console.log("\nđź‘‹ Session ended.");
    process.exit(0);
  });
}

main().catch((err) => {
  console.error("Fatal error:", err);
  process.exit(1);
});
What's happening:
  • We use streamText instead of generateText to stream responses in real-time
  • toolChoice: "auto" allows the model to decide when to use Tave tools
  • stopWhen: stepCountIs(10) allows up to 10 steps for complex multi-tool operations
  • onStepFinish callback displays which tools are being used in real-time
  • We iterate through the text stream to create a typewriter effect as the agent responds
  • The complete response is added to conversation history to maintain context
  • Errors are caught and displayed with helpful retry suggestions

Complete Code

Here's the complete code to get you started with Tave and Vercel AI SDK:

typescript
import "dotenv/config";
import { openai } from "@ai-sdk/openai";
import { Composio } from "@composio/core";
import * as readline from "readline";
import { streamText, type ModelMessage, stepCountIs } from "ai";
import { createMCPClient } from "@ai-sdk/mcp";

const composioAPIKey = process.env.COMPOSIO_API_KEY;
const composioUserID = process.env.COMPOSIO_USER_ID;

if (!process.env.OPENAI_API_KEY) throw new Error("OPENAI_API_KEY is not set");
if (!composioAPIKey) throw new Error("COMPOSIO_API_KEY is not set");
if (!composioUserID) throw new Error("COMPOSIO_USER_ID is not set");

const composio = new Composio({
  apiKey: composioAPIKey,
});

async function main() {
  // Create a tool router session for the user
  const session = await composio.create(composioUserID!, {
    toolkits: ["tave"],
  });

  const mcpUrl = session.mcp.url;

  const mcpClient = await createMCPClient({
    transport: {
      type: "http",
      url: mcpUrl,
      headers: session.mcp.headers, // Authentication headers for the Composio MCP server
    },
  });

  const tools = await mcpClient.tools();

  let messages: ModelMessage[] = [];

  console.log("Chat started! Type 'exit' or 'quit' to end the conversation.\n");
  console.log(
    "Ask any questions related to tave, like summarize my last 5 emails, send an email, etc... :)))\n",
  );

  const rl = readline.createInterface({
    input: process.stdin,
    output: process.stdout,
    prompt: "> ",
  });

  rl.prompt();

  rl.on("line", async (userInput: string) => {
    const trimmedInput = userInput.trim();

    if (["exit", "quit", "bye"].includes(trimmedInput.toLowerCase())) {
      console.log("\nGoodbye!");
      rl.close();
      process.exit(0);
    }

    if (!trimmedInput) {
      rl.prompt();
      return;
    }

    messages.push({ role: "user", content: trimmedInput });
    console.log("\nAgent is thinking...\n");

    try {
      const stream = streamText({
        model: openai("gpt-5"),
        messages,
        tools,
        toolChoice: "auto",
        stopWhen: stepCountIs(10),
        onStepFinish: (step) => {
          for (const toolCall of step.toolCalls) {
            console.log(`[Using tool: ${toolCall.toolName}]`);
          }
          if (step.toolCalls.length > 0) {
            console.log(""); // Add space after tool calls
          }
        },
      });

      for await (const chunk of stream.textStream) {
        process.stdout.write(chunk);
      }

      console.log("\n\n---\n");

      // Get final result for message history
      const response = await stream.response;
      if (response?.messages?.length) {
        messages.push(...response.messages);
      }
    } catch (error) {
      console.error("\nAn error occurred while talking to the agent:");
      console.error(error);
      console.log(
        "\nYou can try again or restart the app if it keeps happening.\n",
      );
    } finally {
      rl.prompt();
    }
  });

  rl.on("close", async () => {
    await mcpClient.close();
    console.log("\nđź‘‹ Session ended.");
    process.exit(0);
  });
}

main().catch((err) => {
  console.error("Fatal error:", err);
  process.exit(1);
});

Conclusion

You've successfully built a Tave agent using the Vercel AI SDK with streaming capabilities! This implementation provides a powerful foundation for building AI applications with natural language interfaces and real-time feedback.

Key features of this implementation:

  • Real-time streaming responses for a better user experience with typewriter effect
  • Live tool execution feedback showing which tools are being used as the agent works
  • Dynamic tool loading through Composio's Tool Router with secure authentication
  • Multi-step tool execution with configurable step limits (up to 10 steps)
  • Comprehensive error handling for robust agent execution
  • Conversation history maintenance for context-aware responses

You can extend this further by adding custom error handling, implementing specific business logic, or integrating additional Composio toolkits to create multi-app workflows.

How to build Tave MCP Agent with another framework

OpenAI Agents SDK

OpenAI Agents SDK

Use Tave MCP with OpenAI Agents SDK

Claude Agent SDK

Claude Agent SDK

Use Tave MCP with Claude Agent SDK

Google ADK

Google ADK

Use Tave MCP with Google ADK

LangChain

LangChain

Use Tave MCP with LangChain

Mastra AI

Mastra AI

Use Tave MCP with Mastra AI

LlamaIndex

LlamaIndex

Use Tave MCP with LlamaIndex

CrewAI

CrewAI

Use Tave MCP with CrewAI

Claude Code

Claude Code

Use Tave MCP with Claude Code

Codex

Codex

Use Tave MCP with Codex

FAQ

What are the differences in Tool Router MCP and Tave MCP?

With a standalone Tave MCP server, the agents and LLMs can only access a fixed set of Tave tools tied to that server. However, with the Composio Tool Router, agents can dynamically load tools from Tave and many other apps based on the task at hand, all through a single MCP endpoint.

Can I use Tool Router MCP with Vercel AI SDK v6?

Yes, you can. Vercel AI SDK v6 fully supports MCP integration. You get structured tool calling, message history handling, and model orchestration while Tool Router takes care of discovering and serving the right Tave tools.

Can I manage the permissions and scopes for Tave while using Tool Router?

Yes, absolutely. You can configure which Tave scopes and actions are allowed when connecting your account to Composio. You can also bring your own OAuth credentials or API configuration so you keep full control over what the agent can do.

How safe is my data with Composio Tool Router?

All sensitive data such as tokens, keys, and configuration is fully encrypted at rest and in transit. Composio is SOC 2 Type 2 compliant and follows strict security practices so your Tave data and credentials are handled as safely as possible.

Used by agents from

Context
Letta
glean
HubSpot
Agent.ai
Altera
DataStax
Entelligence
Rolai
Context
Letta
glean
HubSpot
Agent.ai
Altera
DataStax
Entelligence
Rolai
Context
Letta
glean
HubSpot
Agent.ai
Altera
DataStax
Entelligence
Rolai

Never worry about agent reliability

We handle tool reliability, observability, and security so you never have to second-guess an agent action.

Book a call↗Get started for free