How to integrate Mixmax MCP with Vercel AI SDK v6

Framework Integration Gradient
Mixmax Logo
Vercel AI SDK Logo
divider

Introduction

This guide walks you through connecting Mixmax to Vercel AI SDK v6 using the Composio tool router. By the end, you'll have a working Mixmax agent that can get all mixmax emails tracked today, send a mixmax sequence to new leads, list mixmax meetings scheduled for next week through natural language commands.

This guide will help you understand how to give your Vercel AI SDK agent real control over a Mixmax account through Composio's Mixmax 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 Mixmax integration
  • Using Composio's Tool Router to dynamically load and access Mixmax 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 Mixmax MCP server, and what's possible with it?

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

Supported Tools & Triggers

Tools
Add Contact to GroupTool to add a contact to a contact group.
Add Email to Unsubscribe ListTool to add an email address to the Mixmax unsubscribe list.
Add Snippet to TagTool to add a snippet to a snippet tag in Mixmax.
Add Team MembersTool to add one or more members to a Mixmax team by email.
Cancel Sequence RecipientsTool to cancel specific recipients across all sequences.
Create Code SnippetTool to create a new Formatted Code enhancement in Mixmax.
Create ContactTool to create a new contact in Mixmax.
Create Contact GroupTool to create a new contact group in Mixmax.
Create Contact NoteTool to create a note for a contact.
Create insights reportTool to create a new insights report in Mixmax.
Create Integration EnhancementTool to create a new enhancement integration in Mixmax.
Create Integrations CommandTool to create a new slash command that will be visible to only your account.
Create Integrations SidebarsTool to create a new sidebar integration in Mixmax.
Create Link ResolverTool to create a new link resolver integration in Mixmax.
Create Livefeed SearchTool to create a new saved search in the Mixmax livefeed.
Create MessageTool to create a draft Mixmax message (email).
Create RuleTool to create a new Mixmax rule/webhook.
Create Rule ActionTool to create a new action for a Mixmax rule.
Create Sequence FolderTool to create a new sequence folder in Mixmax.
Create Snippet TagTool to create a new snippet tag (folder) in Mixmax.
Create TeamTool to create a new team in Mixmax.
Delete ContactTool to delete a contact by ID.
Delete Contact GroupTool to delete a contact group by ID.
Delete Contact NoteTool to delete a note from a contact by its ID.
Delete Integration EnhancementTool to delete an integration enhancement by its ID.
Delete slash commandTool to delete a slash command by its unique ID from Mixmax integrations.
Delete Integrations SidebarsTool to delete a sidebar integration by its ID.
Delete Link ResolverTool to permanently delete a link resolver by its ID.
Delete RuleTool to delete a Mixmax rule by its ID.
Delete Rule ActionTool to delete an action from a Mixmax rule.
Delete Sequence FolderTool to permanently delete a sequence folder by its ID.
Delete SnippetTool to delete a snippet by moving it to trash.
Delete Snippet TagTool to delete a snippet tag by its unique ID.
Delete TeamTool to delete a team by its ID.
Delete UnsubscribeTool to remove an email address from the Mixmax unsubscribe list.
Get Code SnippetTool to fetch a single Formatted Code enhancement by its unique ID.
Get contact by IDTool to retrieve a single contact by ID from Mixmax.
Get Contact GroupTool to retrieve a single contact group by ID.
Get Contactgroup ContactsTool to retrieve membership for a contact group in Mixmax.
Get Contact NotesTool to retrieve notes associated with a specific contact.
Get File RequestsTool to retrieve all file requests that you've sent out through Mixmax.
Get insights report by IDTool to retrieve a single insights report by ID from Mixmax.
Get Integrations CommandsTool to retrieve a list of Slash Commands that you have access to in Mixmax.
Get Integrations EnhancementsTool to retrieve a list of enhancements that you have access to in Mixmax.
Get Link Resolver IntegrationsTool to retrieve all link resolver integrations configured in Mixmax.
Get Integrations SidebarsTool to retrieve a list of sidebar integrations that you have access to in Mixmax.
Get live feed activity dataTool to retrieve live feed activity data from Mixmax.
Get Livefeed EventsTool to retrieve events for a specific message in the live feed.
Get Meeting InvitesTool to retrieve a paginated list of meeting invites from Mixmax.
Get Message by IDTool to retrieve a single message by its unique ID.
Get MessagesTool to retrieve messages (emails) from Mixmax.
Get PollsTool to retrieve polls created in Mixmax along with voting information.
Get Q&A SurveysTool to retrieve Question & Answer surveys data from Mixmax.
Get Reports Data TableTool to query analytics data for messages and meetings from Mixmax reports.
Get all rulesTool to retrieve all Mixmax rules for the authenticated user.
Get Rule by IDTool to fetch a single rule by its unique ID from Mixmax.
Get Salesforce account informationTool to retrieve Salesforce account information for the authenticated user.
Get Sequence Folder by IDTool to retrieve a single sequence folder by its ID from Mixmax.
Get Snippet by IDTool to fetch a single snippet by its unique ID from Mixmax.
Get Snippet TagTool to retrieve a single snippet tag (folder) by its unique ID.
Get Snippet TagsTool to retrieve a list of snippet tags (folders) from Mixmax.
Get Team MembersTool to retrieve all members of a Mixmax team.
Get TeamsTool to retrieve all teams that the authenticated user is a member of.
Get UnsubscribesTool to retrieve a list of unsubscribed email addresses from Mixmax.
Get Current User PreferencesTool to retrieve preferences for the currently authenticated user.
Get Current UserTool to retrieve information about the currently authenticated user.
Get Yes/No QuestionsTool to retrieve Yes/No questions created in Mixmax.
List Code SnippetsTool to list all Formatted Code enhancements you've created.
List Contact GroupsTool to retrieve a list of contact groups from Mixmax.
List ContactsTool to retrieve all Mixmax contacts (people you've emailed using Mixmax).
List insights reportsTool to retrieve all insights reports from Mixmax.
List Rule ActionsTool to list all actions for a given Mixmax rule.
List Sequence FoldersTool to retrieve a list of all sequence folders from Mixmax.
List SequencesTool to retrieve a list of sequences from Mixmax.
List SnippetsTool to list templates or snippets that you have access to (including those shared with you).
Update Rule by IDTool to update a rule by its unique ID in Mixmax.
Update Rule ActionTool to update an action within a Mixmax rule.
Update Snippet by IDTool to update a snippet by its unique ID in Mixmax.
Post Reports Data TableTool to generate a table report with specified analytics data for messages, meetings, or sequences.
Query ContactsTool to search for matching contacts across multiple sources including Mixmax, Google Directory, and Salesforce (contacts, leads, accounts, and opportunities).
Remove Contact from GroupTool to remove a contact from a contact group.
Remove Team MemberTool to remove a member from a Mixmax team.
Search SequencesTool to search sequence recipients by email and sequenceId from Mixmax.
Send Message DraftTool to send a message draft by its unique ID.
Send Snippet as EmailTool to send a snippet as an email to specified recipients.
Send Test MessageTool to send a test Mixmax message to the authenticated user, resolving any variables.
Update Appointment Link Calendar NameTool to update your custom "Mixmax Calendar" vanity URL name.
Update ContactTool to update a specific contact by ID.
Update Contact GroupTool to update a contact group by ID.
Update Contact NoteTool to update a note on a contact by its ID.
Update insights reportTool to update an existing insights report in Mixmax.
Update Message by IDTool to update a message by its unique ID.
Update Sequence Folder by IDTool to update a sequence folder by its unique ID in Mixmax.
Update Snippet TagTool to update a snippet tag (folder) by its unique ID.
Update TeamTool to update a team's information by ID.
Update Current User PreferencesTool to update preferences for the currently authenticated user.

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: ["mixmax"],
  });

  const mcpUrl = session.mcp.url;
What's happening:
  • We're creating a Tool Router session that gives your agent access to Mixmax 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 Mixmax-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 Mixmax 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 mixmax, 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 Mixmax 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 Mixmax 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: ["mixmax"],
  });

  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 mixmax, 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 Mixmax 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 Mixmax MCP Agent with another framework

FAQ

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

With a standalone Mixmax MCP server, the agents and LLMs can only access a fixed set of Mixmax tools tied to that server. However, with the Composio Tool Router, agents can dynamically load tools from Mixmax 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 Mixmax tools.

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

Yes, absolutely. You can configure which Mixmax 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 Mixmax 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.