How to integrate Apollo MCP with Vercel AI SDK v6

Connect Vercel AI SDK v6 to Apollo MCP. Bulk enrich profiles for new leads, add contacts to outreach sequence now, and more using natural language, with authentication handled for you.

Get started for free Get a demo

Apollo

Api Key

Apollo is a CRM and lead generation platform that helps businesses discover contacts and manage sales pipelines. Use it to streamline customer outreach and track your deals from one place.

67 Tools

Managed auth

Connect Apollo without auth hassles

We manage OAuth, API keys, token refresh, and scopes — you just build.

Book a call

Explore the MCP Gateway

Introduction

This guide walks you through connecting Apollo to Vercel AI SDK v6 using the Composio tool router. By the end, you'll have a working Apollo agent that can bulk enrich profiles for new leads, add contacts to outreach sequence now, create a new sales deal for acme through natural language commands.

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

The Apollo MCP server is an implementation of the Model Context Protocol that connects your AI agent and assistants like Claude, Cursor, etc directly to your Apollo account. It provides structured and secure access to your contacts, organizations, outreach, pipeline, conversations, and analytics, so your agent can find prospects, enrich records, manage sequence participation, update sales stages, and review performance on your behalf.

Prospect search and enrichment: Have your agent search for people and enrich person or organization records with data from Apollo.
Sequence and outreach management: Let the agent find sequences, add contacts to them, update contact participation status, and review sent outreach emails and engagement activity.
Contact and account updates: Direct your agent to update contact stages, account stages, and ownership assignments across your sales records.
Pipeline, task, and conversation research: Instruct your agent to retrieve deal details, search sales tasks, and review recorded conversations with transcripts, participants, and related opportunities.
Sales insights and company signals: Have your agent query analytics reports, check credit usage, find company news, and review organization job postings.

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:

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

Step-by-step Guide

Step by step09 STEPS

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: ["apollo"],
  });

  const mcpUrl = session.mcp.url;

What's happening:

We're creating a Tool Router session that gives your agent access to Apollo 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 Apollo-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 Apollo 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 apollo, 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("\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 Apollo 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 Apollo 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: ["apollo"],
  });

  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 apollo, 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 Apollo 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 Apollo action and event your agent gets out of the box.

Activate Sequence

Activate an inactive Apollo sequence and begin its scheduled outreach.

Add Contacts to Sequence

Adds contacts to a specified Apollo email sequence and returns the contact details.

Archive Sequence

Archive an Apollo sequence, making it inactive and finishing all enrolled contacts.

Bulk organization enrichment

Enriches data for up to 10 organizations simultaneously by providing a list of their base company domains (e.

Bulk people enrichment

Use to enrich multiple person profiles simultaneously with comprehensive data from Apollo's database.

Bulk update account stage

Bulk updates the stage for specified existing Apollo.

Complete Apollo Task

Permanently mark an Apollo task as completed and return its resulting state.

Create an Apollo account

Creates a new account in Apollo.

Bulk create Apollo accounts

Creates multiple accounts in Apollo.

Bulk create Apollo contacts

Tool to bulk create multiple contacts in Apollo with a single API call.

Create call record in Apollo

Tool to log call records in Apollo from external systems.

Create Apollo contact

Creates a new contact in Apollo.

Create custom field

Creates a new custom field in Apollo.

Create Apollo deal

Creates a new sales opportunity (deal) in Apollo.

Create Email Draft

Create an unsent Apollo outreach draft, or return the existing draft already linked to outreach_task_id.

Create Apollo Sequence

Create an Apollo outreach sequence, optionally including steps and email variants.

Create Apollo Task

Tool to create a single task in Apollo.

Deactivate Sequence

Deactivate an active Apollo sequence.

Export Conversations

Start an asynchronous gzipped JSON conversation export and email a team member when it is ready.

Get Account by ID

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

Check Apollo API key status

Tool to check whether the provided Apollo API key is valid and accepted by Apollo (health/auth check).

Get Apollo Contact

Retrieves detailed information about a specific contact by its ID.

Get Conversation

Retrieve full Apollo conversation details, including transcript, participants, recordings, and opportunities.

Get Conversation Export

Retrieve the download redirect URL for a previously started Apollo conversation export.

Get Apollo deal

Retrieves information about a specific deal by its ID.

Get Email Activities

Retrieves content, open and click activity, and recipient details for one Apollo sequence email.

Get Email Content

Retrieve content for up to 10 sent team outreach emails.

Get Labels

Retrieves all labels from Apollo.

Get opportunity stages

Retrieves all configured opportunity (deal) stages from the Apollo.

Get Organization by ID

Retrieves complete information about a specific organization by its Apollo ID.

Get Organization Job Postings

Retrieves paginated job postings for a specified organization by its ID, optionally filtering by domain; ensure `organization_id` is a valid identifier.

Get Apollo Task

Retrieves the full details of one Apollo task by ID, including associated records and type-specific data when present.

Get typed custom fields

Retrieves all typed custom field definitions available in the Apollo.

List Apollo account stages

Retrieves the IDs for all available account stages in your team's Apollo account.

List apollo contact stages

Retrieves all available contact stages from an Apollo account, including their unique IDs and names.

List Apollo deals

Retrieves a list of deals from Apollo, using Apollo's default sort order if 'sort_by_field' is omitted.

List email accounts

Retrieves all email accounts and their details for the authenticated user; takes no parameters.

List Fields

Retrieves all field definitions from Apollo.

List Apollo Notes

List notes associated with Apollo records.

List Apollo Users

Retrieves a list of all users (teammates) associated with the Apollo account, supporting pagination via `page` and `per_page` parameters.

Enrich organization data

Fetches comprehensive organization enrichment data from Apollo.

Search organizations in Apollo

Searches Apollo's database for organizations using various filters; consumes credits on every call (unavailable on free plans) — avoid re-running identical queries and surface quota errors rather than retrying.

Enrich person with Apollo

Enriches and retrieves information for a person from Apollo.

Apollo people search

Searches Apollo's contact database for people using various filters; results capped at 50,000 records and does not enrich contact data.

Query Analytics Report

Query Apollo analytics metrics, dimensions, filters, and date ranges.

Search Apollo Accounts

Searches for accounts within your existing Apollo.

Search for Calls

Searches for call records in Apollo.

Search Apollo contacts

Searches Apollo contacts using keywords, stage IDs (from 'List Contact Stages' action), or sorting (max 50,000 records; `sort_ascending` requires `sort_by_field`).

Search Apollo Conversations

Search Apollo video meetings and dialer calls using record, participant, date, scorecard, sorting, and visibility filters.

Search news articles

Tool to search for news articles about companies in Apollo's database.

Search outreach emails

Tool to search for outreach emails sent through Apollo sequences.

Search sequences

Searches for sequences (e.

Search tasks

Searches for tasks in Apollo.

Send Email Now

Queue an Apollo email for external sending; success means queued, not delivered.

Update an Apollo account

Updates specified attributes of an existing account in Apollo.

Update account ownership

Updates the ownership of multiple Apollo accounts to a specified user.

Update Apollo call record

Tool to update an existing call record in Apollo.

Update Apollo contact details

Tool to update an existing contact's information in Apollo.

Update contact ownership

Updates the ownership of specified Apollo contacts to a given Apollo user, who must be part of the same team.

Bulk update Apollo contacts

Tool to bulk update multiple Apollo contacts with a single API call.

Update contact stage

Updates the stage for one or more existing contacts in Apollo.

Update contact status in sequence

Updates a contact's status within a designated Apollo sequence, but cannot set the status to 'active'.

Update Apollo deal

Updates specified fields of an existing Apollo.

Update Sequence

Update an Apollo sequence's settings, ownership, sharing, steps, and touches.

Update Apollo Task

Update editable details of an existing Apollo task.

View API Usage Stats

Fetches Apollo API usage statistics and rate limits for the connected team.

View Credit Usage Stats

Read team balances by credit type for Apollo's current billing cycle.

FRAMEWORKS

How to build Apollo MCP Agent with another framework

Claude Code

Use Apollo MCP with Claude Code

Cursor

Use Apollo MCP with Cursor

ChatGPT

Use Apollo MCP with ChatGPT

Claude Cowork

Use Apollo MCP with Claude Cowork

Antigravity

Use Apollo MCP with Antigravity

OpenAI Agents SDK

Use Apollo MCP with OpenAI Agents SDK

Claude Agent SDK

Use Apollo MCP with Claude Agent SDK

Grok Bot

Use Apollo MCP with Grok Bot

Muse

Use Apollo MCP with Muse

Codex

Use Apollo MCP with Codex

Grok Build

Use Apollo MCP with Grok Build

VS Code

Use Apollo MCP with VS Code

OpenCode

Use Apollo MCP with OpenCode

OpenClaw

Use Apollo MCP with OpenClaw

Hermes

Use Apollo MCP with Hermes

Atomic Agent

Use Apollo MCP with Atomic Agent

Google ADK

Use Apollo MCP with Google ADK

LangChain

Use Apollo MCP with LangChain

Mastra AI

Use Apollo MCP with Mastra AI

LlamaIndex

Use Apollo MCP with LlamaIndex

CrewAI

Use Apollo MCP with CrewAI

Pydantic AI

Use Apollo MCP with Pydantic AI

AutoGen

Use Apollo MCP with AutoGen

MORE TOOLKITS

Explore Other Toolkits

Toolkit marketplace

HubSpot

Oauth2Api Key

HubSpot is an all-in-one marketing, sales, and customer service platform. It lets teams nurture leads, automate outreach, and track every customer interaction in one place.

Pipedrive

Oauth2Api Key

Pipedrive is a sales management platform offering pipeline visualization, lead tracking, and workflow automation. It helps sales teams keep deals moving forward efficiently and never miss a follow-up.

Salesforce

Oauth2S2s Oauth2

Salesforce is a leading CRM platform that helps businesses manage sales, service, and marketing. It centralizes customer data, enabling teams to drive growth and build strong relationships.

Attio

Oauth2

Attio is a customizable CRM and workspace for managing your team's relationships and workflows. It helps teams organize contacts, automate tasks, and collaborate more efficiently.

FAQ

Frequently asked questions

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

Yes, absolutely. You can configure which Apollo 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 Apollo data and credentials are handled as safely as possible.

Start with Apollo.It takes 30 seconds.

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

Start building

How to integrate Apollo MCP with Vercel AI SDK v6

Connect Apollo without auth hassles

Introduction

Also integrate Apollo with

TL;DR

What is Vercel AI SDK?

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

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

What is Composio SDK?

How the Composio SDK works

Step-by-step Guide

Prerequisites

Getting API Keys for OpenAI and Composio

Install required dependencies

Set up environment variables

Import required modules and validate environment

Create Tool Router session and initialize MCP client

Connect to MCP server and retrieve tools

Initialize conversation and CLI interface

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

Complete Code

Conclusion

Supported Tools

How to build Apollo MCP Agent with another framework

Claude Code

Cursor

ChatGPT

Claude Cowork

Antigravity

OpenAI Agents SDK

Claude Agent SDK

Grok Bot

Muse

Codex

Grok Build

VS Code

OpenCode

OpenClaw

Hermes

Atomic Agent

Google ADK

LangChain

Mastra AI

LlamaIndex

CrewAI

Pydantic AI

AutoGen

Explore Other Toolkits

HubSpot

Pipedrive

Salesforce

Attio

Frequently asked questions

What are the differences in Tool Router MCP and Apollo MCP?+

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

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

How safe is my data with Composio Tool Router?+

Start with Apollo.It takes 30 seconds.