How to integrate Intercom MCP with Vercel AI SDK

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

Introduction

This guide walks you through connecting Intercom to Vercel AI SDK using the Composio tool router. By the end, you'll have a working Intercom agent that can add tag 'vip' to contact john doe, assign open conversation #123 to support team, create note for contact emily about refund, close all resolved conversations from today through natural language commands.

This guide will help you understand how to give your Vercel AI SDK agent real control over a Intercom account through Composio's Intercom 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 Intercom integration
  • Using Composio's Tool Router to dynamically load and access Intercom 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
  • Step Counting: Control multi-step tool execution
  • OpenAI Provider: Native integration with OpenAI models

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

The Intercom MCP server is an implementation of the Model Context Protocol that connects your AI agent and assistants like Claude, Cursor, etc directly to your Intercom account. It provides structured and secure access to your customer engagement platform, so your agent can perform actions like managing conversations, tagging contacts, creating articles, and updating company records on your behalf.

  • Conversation management and assignment: Let your agent assign conversations to teams or admins, create new conversations, and close them when resolved, streamlining your support workflow.
  • Contact tagging and note creation: Effortlessly tag contacts with relevant labels or add detailed notes for context, making customer follow-ups more organized and actionable.
  • Automated company and contact updates: Enable your agent to attach contacts to companies, create or update company records, and keep your Intercom data clean and up to date.
  • Article and collection creation: Let your agent publish new articles or create help center collections to expand your self-serve support resources without manual effort.
  • Subscription and message preferences management: Allow your agent to add or manage subscriptions for contacts, helping you personalize communication and respect user preferences automatically.

Supported Tools & Triggers

Tools
Add subscription to a contactYou can add a specific subscription to a contact.
Add tag to a contactYou can tag a specific contact.
Assign conversationAssigns a conversation to a specific admin or team in intercom
Attach a contact to a companyYou can attach a company to a single contact.
Close conversationCloses a conversation in intercom, marking it as resolved
Create a collectionYou can create a new collection by making a post request to `https://api.
Create an articleYou can create a new article by making a post request to `https://api.
Create a noteYou can add a note to a single contact.
Create conversationCreates a new conversation in intercom
Create or update a companyYou can create or update a company.
Delete a collectionYou can delete a single collection by making a delete request to `https://api.
Delete a companyYou can delete a single company.
Delete a contactYou can delete a single contact.
Delete an articleYou can delete a single article by making a delete request to `https://api.
Detach a contact from a companyYou can detach a company from a single contact.
Get a contactYou can fetch the details of a single contact.
Get conversationRetrieves a specific conversation by id with all messages and details
Identify an adminYou can view the currently authorised admin along with the embedded app object (a "workspace" in legacy terminology).
List all activity logsYou can get a log of activities by all admins in an app.
List all adminsYou can fetch a list of admins for a given workspace.
List all articlesYou can fetch a list of all articles by making a get request to `https://api.
List all collectionsYou can fetch a list of all collections by making a get request to `https://api.
List all companiesYou can list companies.
List all help centersYou can list all help centers by making a get request to `https://api.
List all notesYou can fetch a list of notes that are associated to a contact.
List attached companies for contactYou can fetch a list of companies that are associated to a contact.
List attached contactsYou can fetch a list of all contacts that belong to a company.
List attached segments for companiesYou can fetch a list of all segments that belong to a company.
List attached segments for contactYou can fetch a list of segments that are associated to a contact.
List conversationsLists conversations from intercom with filtering and pagination support
List subscriptions for a contactYou can fetch a list of subscription types that are attached to a contact.
List tags attached to a contactYou can fetch a list of all tags that are attached to a specific contact.
Merge a lead and a userYou can merge a contact with a `role` of `lead` into a contact with a `role` of `user`.
Remove subscription from a contactYou can remove a specific subscription from a contact.
Remove tag from a contactYou can remove tag from a specific contact.
Reopen conversationReopens a closed conversation in intercom
Reply to conversationSends a reply to an existing conversation in intercom
Retrieve a collectionYou can fetch the details of a single collection by making a get request to `https://api.
Retrieve a company by idYou can fetch a single company.
Retrieve a help centerYou can fetch the details of a single help center by making a get request to `https://api.
Retrieve an adminYou can retrieve the details of a single admin.
Retrieve an articleYou can fetch the details of a single article by making a get request to `https://api.
Retrieve companiesYou can fetch a single company by passing in `company id` or `name`.
Scroll over all companiesThe `list all companies` functionality does not work well for huge datasets, and can result in errors and performance problems when paging deeply.
Search conversationsSearches for conversations using query string with support for filtering and sorting
Search for articlesYou can search for articles by making a get request to `https://api.
Set an admin to awayYou can set an admin as away for the inbox.
Update a collectionYou can update the details of a single collection by making a put request to `https://api.
Update a companyYou can update a single company using the intercom provisioned `id`.
Update a contactYou can update an existing contact (ie.
Update an articleYou can update the details of a single article by making a put request to `https://api.

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 { experimental_createMCPClient as 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: ["intercom"],
  });

  const mcpUrl = session.mcp.url;
What's happening:
  • We're creating a Tool Router session that gives your agent access to Intercom 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 Intercom-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 Intercom 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 intercom, 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("\nšŸ‘‹ 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 Intercom 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 Intercom 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 { experimental_createMCPClient as 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: ["intercom"],
  });

  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 intercom, 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("\nšŸ‘‹ Session ended.");
    process.exit(0);
  });
}

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

Conclusion

You've successfully built a Intercom 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 Intercom MCP Agent with another framework

OpenAI Agents SDK

OpenAI Agents SDK

Use Intercom MCP with OpenAI Agents SDK

Claude Agent SDK

Claude Agent SDK

Use Intercom MCP with Claude Agent SDK

Google ADK

Google ADK

Use Intercom MCP with Google ADK

LangChain

LangChain

Use Intercom MCP with LangChain

Mastra AI

Mastra AI

Use Intercom MCP with Mastra AI

LlamaIndex

LlamaIndex

Use Intercom MCP with LlamaIndex

CrewAI

CrewAI

Use Intercom MCP with CrewAI

Claude Code

Claude Code

Use Intercom MCP with Claude Code

FAQ

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

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

Yes, you can. Vercel AI SDK 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 Intercom tools.

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

Yes, absolutely. You can configure which Intercom 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 Intercom 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 callā†—Get started for free