How to integrate Heygen MCP with Vercel AI SDK v6

This guide walks you through connecting Heygen to Vercel AI SDK v6 using the Composio tool router. By the end, you'll have a working Heygen agent that can create a talking photo from your selfie, add new contacts to your video project, list all available streaming avatars for today through natural language commands. This guide will help you understand how to give your Vercel AI SDK agent real control over a Heygen account through Composio's Heygen MCP server. Before we dive in, let's take a quick look at the key ideas and tools involved.

Get started for freeGet a demo
Heygen logoHeygen
Api Key

HeyGen is an AI-driven video creation platform that streamlines content production. Create, edit, and automate video workflows with generative AI for faster results.

71 Tools

Introduction

This guide walks you through connecting Heygen to Vercel AI SDK v6 using the Composio tool router. By the end, you'll have a working Heygen agent that can create a talking photo from your selfie, add new contacts to your video project, list all available streaming avatars for today through natural language commands.

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

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

Also integrate Heygen with

ChatGPTOpenAI Agents SDKClaude Agent SDKClaude CodeClaude CoworkCodexOpenCodeCursorVS CodeOpenClawHermesCLIGoogle ADKLangChainMastra AILlamaIndexCrewAI

TL;DR

Here's what you'll learn:
  • How to set up and configure a Vercel AI SDK agent with Heygen integration
  • Using Composio's Tool Router to dynamically load and access Heygen 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 Heygen MCP server, and what's possible with it?

The Heygen MCP server is an implementation of the Model Context Protocol that connects your AI agent and assistants like Claude, Cursor, etc directly to your Heygen account. It provides structured and secure access to your Heygen video platform, so your agent can perform actions like creating talking photos, managing video assets, personalizing video projects, and controlling streaming avatars on your behalf.

  • AI-powered video asset creation: Add new avatars, backgrounds, or other content elements directly into your Heygen library for use in video generation workflows.
  • Personalized video project management: Let your agent add contacts to personalized video projects and fetch detailed project information to streamline large-scale video personalization efforts.
  • Dynamic talking photo generation: Instantly create engaging talking photos from uploaded images, transforming static pictures into interactive, speaking content for marketing or educational use.
  • Real-time avatar streaming control: Retrieve a list of streaming avatars, generate streaming tokens, and manage live avatar sessions—including interrupting or monitoring ongoing streams for responsive, interactive experiences.
  • Audience insights and analytics: Fetch detailed audience engagement and preference data for personalized video campaigns, helping you optimize your content and targeting strategies.

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

What is Composio SDK?

Composio's Composio SDK 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 Composio SDK

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

How the Composio SDK works

The Composio SDK 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

Step by step09 STEPS
1

Prerequisites

Before you begin, make sure you have:
  • Node.js and npm installed
  • A Composio account with API key
  • An OpenAI API key
2

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.
3

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
4

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
5

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
6

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: ["heygen"],
  });

  const mcpUrl = session.mcp.url;
What's happening:
  • We're creating a Tool Router session that gives your agent access to Heygen 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 Heygen-related tools through the MCP protocol
7

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 Heygen tools that the agent can use
8

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 heygen, 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
9

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("\n👋 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 Heygen 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 Heygen 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: ["heygen"],
  });

  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 heygen, 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("\n👋 Session ended.");
    process.exit(0);
  });
}

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

Conclusion

You've successfully built a Heygen 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.
TOOLS

Supported Tools

Every Heygen action and event your agent gets out of the box.

Add looks to photo avatar group

Tool to add additional looks (images) to an existing photo avatar group.

Add motion to photo avatar

Tool to add natural motion or animation to an existing photo avatar, turning a still image into a moving, lifelike one.

Check photo/look generation status

Tool to retrieve the current status and details of a photo avatar or look generation by its ID.

Create a new folder

Tool to create a new folder under your HeyGen account.

Create knowledge base

Tool to create a new knowledge base with a specified name, opening line, and custom prompt for Interactive Avatar streaming sessions.

Create photo avatar group

Tool to create an avatar group which can include both AI-generated and user-uploaded photos of the avatar.

Create WebM video

Tool to create a WebM format video with transparent background featuring an avatar speaking text or audio.

Delete an avatar video

Tool to delete a video by its ID from HeyGen account.

Delete asset

Tool to delete a specific asset by its unique asset ID from HeyGen account.

Delete knowledge base

Tool to delete a specific knowledge base by its ID.

Delete photo avatar

Tool to delete a photo avatar by its ID from HeyGen account.

Delete photo avatar group

Tool to delete a photo avatar group by its ID from HeyGen account.

Delete webhook endpoint

Tool to permanently delete a webhook endpoint from HeyGen.

Generate photo avatar photos

Tool to generate a personalized AI avatar based on text prompts and attribute inputs.

Generate text-to-speech audio

Tool to generate a speech audio file from any text input using a specified voice, powered by HeyGen's Starfish model.

Generate voice audio preview

Tool to generate a short audio preview with the specified voice and text.

Get current user information

Tool to retrieve the profile information of the currently authenticated user.

Get training job status

Tool to retrieve the current training status and details of a photo avatar training job by group ID.

Keep alive streaming session

Tool to reset the idle timeout counter for an active streaming session.

List all avatar groups

Tool to retrieve a list of all available avatar groups in your HeyGen account.

List all locales for voices

Tool to retrieve available locales for multilingual voices.

List assets

Tool to retrieve a paginated list of all assets (images, audios, videos) created under your account.

List brand voices

Tool to retrieve a list of brand glossaries created under your account.

List folders

Tool to retrieve a paginated list of folders created under your account.

List avatars in a specific group

Tool to retrieve all avatars that belong to a specific avatar group.

List knowledge bases

Tool to retrieve a list of all existing knowledge bases associated with your account.

List session history

Tool to retrieve a paginated history of all streaming sessions with metadata including session duration, timestamps, and other details.

List streaming avatars

Tool to retrieve a list of public and custom interactive avatars for streaming.

List TTS voices

Tool to retrieve a list of public and custom voices compatible with HeyGen's Starfish TTS model.

Create new streaming session

Tool to initiate a new streaming session with an Interactive Avatar for real-time interactions.

Add contact to personalized video project

This endpoint allows you to add one or more contacts to a specific personalized video project in the Heygen platform.

Fetch personalized video project detail

Retrieves detailed information about a specific personalized video project in HeyGen, including status, settings, and metadata.

Get photo avatar details

Tool to retrieve detailed information about a photo avatar or look using its unique ID.

Post talking photo binary image

Creates a talking photo by processing an uploaded image file.

Restore folder from trash

Tool to restore a previously deleted folder from trash by its unique folder ID.

Retrieve avatar details

Tool to retrieve detailed information about a specific avatar by its ID.

Retrieve sharable video URL

Tool to generate a public, shareable URL for a specific video by ID.

Retrieve template details v3

Tool to retrieve comprehensive details about a specific template including all available variables for replacement and scene mappings for templates built in the New AI Studio.

Retrieve video status details

Tool to retrieve the current processing status and metadata for a specific video by ID.

Search public avatar groups

Tool to search and retrieve public avatar groups with optional filters and pagination.

Create streaming token with expiry

Creates a new streaming token for use with HeyGen's real-time avatar and voice chat services.

Post streaming ice candidates

This endpoint is used to submit ICE (Interactive Connectivity Establishment) candidate information for a specific streaming session in the Heygen platform.

Interrupt streaming session

Interrupts an ongoing streaming session with an interactive avatar.

Retrieve streaming list

Retrieves a list of active or available streaming sessions or content within the HeyGen platform.

Set streaming quality endpoint

Initiates a new streaming session with HeyGen, allowing users to start an avatar-based stream with specified quality settings.

Start streaming session

Initiates a new streaming session for real-time communication in the HeyGen platform.

Stop streaming session

The streaming.

Send task to streaming avatar

The StreamingTask endpoint initiates a real-time speaking task for an AI-driven avatar within an active streaming session.

Move folder to trash

Tool to move a specific folder to trash by its unique folder ID.

Update folder name

Tool to update (rename) an existing folder by its unique folder ID.

Update knowledge base

Tool to update an existing knowledge base by its ID.

Update webhook endpoint

Tool to modify an existing webhook endpoint configuration including URL and subscribed event types.

Upload asset

Tool to upload a file asset (image, video, or audio) to HeyGen platform.

Upscale motion avatar

Tool to enhance the resolution and visual quality of a motion avatar.

List avatars endpoint

Retrieves a list of available avatars from the Heygen platform.

List talking photo entries

Retrieves a list of talking photos created using the HeyGen platform.

List videos

Retrieves a list of videos associated with the user's account on the HeyGen platform.

List voice endpoints

Retrieves a comprehensive list of all available voices in the HeyGen platform.

Add webhook endpoint

Adds a new webhook endpoint to receive real-time notifications for specified Heygen events.

List webhook endpoints

Retrieves a list of all webhook endpoints configured for your HeyGen account.

List webhooks

Retrieves a list of all available webhook event types that can be subscribed to in your HeyGen account.

Delete talking photo by id

Deletes a specific talking photo from the HeyGen platform using its unique identifier.

Generate video template with variables

Generates a customized video from a pre-existing template using HeyGen's AI platform.

Get template by id

Retrieves a specific template from the HeyGen platform using its unique identifier.

Retrieve all templates

Retrieves a list of available avatar templates from the HeyGen platform.

Retrieve user remaining quota

Retrieves the current remaining quota for the authenticated user on the HeyGen platform.

Generate video with inputs

Generates a customized video using Heygen's AI-driven platform.

Post video translate request

The TranslateVideo endpoint enables the translation of video content from one language to another.

Retrieve video translation by id

Retrieves the current status of a video translation job in the HeyGen platform.

Retrieve video translation target languages

Retrieves a list of all available target languages supported by HeyGen's video translation feature.

List available voices

Retrieves a list of available voice models and options that can be used with HeyGen's AI-driven video creation platform.

FRAMEWORKS

How to build Heygen MCP Agent with another framework

ChatGPT logo

ChatGPT

Use Heygen MCP with ChatGPT

OpenAI Agents SDK logo

OpenAI Agents SDK

Use Heygen MCP with OpenAI Agents SDK

Claude Agent SDK logo

Claude Agent SDK

Use Heygen MCP with Claude Agent SDK

Claude Code logo

Claude Code

Use Heygen MCP with Claude Code

Claude Cowork logo

Claude Cowork

Use Heygen MCP with Claude Cowork

Codex logo

Codex

Use Heygen MCP with Codex

Cursor logo

Cursor

Use Heygen MCP with Cursor

OpenClaw logo

OpenClaw

Use Heygen MCP with OpenClaw

Hermes logo

Hermes

Use Heygen MCP with Hermes

CLI logo

CLI

Use Heygen MCP with CLI

Google ADK logo

Google ADK

Use Heygen MCP with Google ADK

LangChain logo

LangChain

Use Heygen MCP with LangChain

Mastra AI logo

Mastra AI

Use Heygen MCP with Mastra AI

LlamaIndex logo

LlamaIndex

Use Heygen MCP with LlamaIndex

CrewAI logo

CrewAI

Use Heygen MCP with CrewAI

OpenCode logo

OpenCode

Use Heygen MCP with OpenCode

VS Code logo

VS Code

Use Heygen MCP with VS Code

FAQ

Frequently asked questions

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

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 Heygen tools.

Yes, absolutely. You can configure which Heygen 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.

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 Heygen data and credentials are handled as safely as possible.

Start with Heygen.It takes 30 seconds.

Managed auth, hosted MCP servers, and every Heygen tool your agent needs.Free to start.

Start building
Customer stories

Teams ship real agents on Composio

GET STARTED FOR FREEBOOK A DEMO
agent.aiZoomLettaGleanHubSpotAWSWabi

Karan skipped his own birthday party to fix our critical issue. It was 10pm and he diverted his Waymo to help us instead. This really sets the bar — it shows the commitment you need when users rely on your software.

Harsha Gaddipati
Harsha Gaddipati
Co-founder, Slashy

A lot of students tell us that the moment their connected tools start talking to each other inside Opennote feels almost magical. The agent just knows them, and it's immensely helped keep new users on the platform.

Abhi Arya
Abhi Arya
Co-founder, Opennote

We chose Composio over Pipedream because it delivered depth where it mattered — niche tools and tricky edge cases other platforms simply ignored. That gave us confidence to scale without compromising.

Nirman Dave
Nirman Dave
CEO, Zams

As a solo builder, shipping fast is life or death. The only way I can outcompete incumbents is by outmanoeuvring them. Getting bogged down managing agent auth would have been a death sentence.

Ryan Yu
Ryan Yu
Founder, Extra Thursday

Before Composio, adding tool integrations was slow and resource-intensive. Each one could take weeks of engineering time, and maintaining them meant constantly keeping up with API changes.

Tomisin Jenrola
Tomisin Jenrola
Founder & CEO, SwarmZero

With hands-on help from their founder, we integrated Gmail and Google Drive in just 30 minutes. This level of personal support and commitment is exactly what startups should strive for.

Jerome Leclanche
Jerome Leclanche
Co-Founder, Ingram Technologies