How to integrate Control d MCP with Vercel AI SDK v6

Trusted by
AWS
Glean
Zoom
Airtable
GET STARTED FOR FREEGET A DEMO

30 min · no commitment · see it on your stack

Control d logo
Vercel AI SDK logo
divider

Introduction

This guide walks you through connecting Control d to Vercel AI SDK v6 using the Composio tool router. By the end, you'll have a working Control d agent that can list all devices connected to your account, remove a device by its id, show known access ips for your network through natural language commands.

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

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

Also integrate Control d with

TL;DR

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

The Control d MCP server is an implementation of the Model Context Protocol that connects your AI agent and assistants like Claude, Cursor, etc directly to your Control d account. It provides structured and secure access to your DNS filtering and device management environment, so your agent can perform actions like managing devices, enforcing policies, retrieving analytics, and monitoring network access on your behalf.

  • Device inventory management: Easily list all devices on your account or remove specific devices by their identifier for streamlined device control.
  • Profile and rule administration: Direct your agent to delete profiles, custom rules, or schedules—helping you maintain and enforce up-to-date network policies.
  • Network access monitoring: Retrieve a list of known access IPs to keep tabs on which endpoints are connecting to your network infrastructure.
  • Analytics endpoints discovery: Quickly fetch available analytics storage regions and endpoints so you can integrate and analyze DNS traffic data efficiently.
  • Organization details access: Have the agent fetch and present your organization's account details for easy reference and auditing.

Supported Tools & Triggers

Tools
Delete Device by IDPermanently delete a Control-D device/endpoint by its ID.
Delete ProfilePermanently deletes a Control D profile by its unique identifier (PK).
Delete Profile Rule by Rule IDDelete a custom DNS rule from a Control D profile by its rule identifier (hostname/domain).
Delete Rule from FolderDelete a custom DNS rule from a specific folder in a Control D profile.
Delete Profile ScheduleTool to delete a specific schedule within a profile.
List Known Access IPsList up to the latest 50 IP addresses that were used to query against a specific Device (resolver).
Get Analytics EndpointsTool to list analytics storage regions and their endpoints.
Get Analytics LevelsTool to retrieve available analytics log levels for Control D devices.
Get Billing PaymentsTool to retrieve billing history of all payments made.
Get Billing ProductsRetrieve all products currently activated on the Control D account.
Get DevicesLists all Control D devices (endpoints) associated with the account.
Get Device TypesList all allowed device types in Control D.
Get IPTool to retrieve the current IP address and datacenter information for the API request.
Get Network StatsTool to retrieve network stats on available services in different POPs (Points of Presence).
Get Organization MembersTool to view organization membership.
Get Organization DetailsTool to view the authenticated organization's details.
Get Sub-OrganizationsTool to view sub-organizations and their details.
Get ProfilesTool to list all profiles associated with the authenticated account.
Get Profile OptionsRetrieves all available configuration options for DNS profiles in Control D.
Get Profile by IDTool to retrieve details of a specific profile by its ID.
Get Profile AnalyticsRetrieve analytics data for a Control D profile.
Get Profile Analytics LogsRetrieves DNS query activity logs for a specific Control D profile.
Get Analytics Log EntryTool to retrieve a specific analytics log entry by its ID.
Get Profile Analytics SummaryTool to fetch a summary of analytics data for a given profile.
Get Profile Analytics Top DomainsTool to fetch top domains accessed within a specific profile.
Get Profile Top ServicesTool to fetch top services accessed within a profile.
Get Profile FiltersList all native (Control D curated) filters for a profile and their current states.
List External Filters for ProfileTool to list third-party filters for a specific profile.
Get Profile FoldersList all rule folders (groups) within a Control D profile.
List Custom DNS Rules for ProfileRetrieve custom DNS rules for a Control D profile.
Get Specific Rule in FolderTool to retrieve a specific rule within a folder by its ID.
Get Profile SchedulesTool to list schedules associated with a specific profile.
Get Profile ScheduleTool to retrieve a specific schedule by its ID within a profile.
Get Profile ServicesTool to list services associated with a specific profile.
Get ProxiesTool to retrieve the list of usable proxy locations that traffic can be redirected through.
Get Service CategoriesList all available service categories in Control D.
List Services by CategoryRetrieves all services within a specific ControlD service category.
Get UsersRetrieve the authenticated user's account information from Control D.
Create DeviceCreate a new device (DNS endpoint) in Control D.
Create ProfileCreate a new blank profile or clone an existing one.
Create Custom DNS RuleCreate custom DNS rules for a profile to control domain resolution.
Create Custom Rules in Profile FolderTool to create custom rules within a specific folder for a profile.
Create Profile ScheduleCreate a new time-based schedule within a Control D profile.
Modify DeviceModify an existing Control D device's settings.
Modify OrganizationModify organization settings such as name, contact details, website, and device limits.
Modify ProfileModify an existing profile by its ID.
Bulk Update Profile FiltersTool to bulk update filters on a specific profile.
Update External Filters for ProfileTool to update external filters for a specific profile.
Modify Profile FilterModify the enabled state of a specific native filter on a profile.
Modify Custom Rule for ProfileModify an existing custom DNS rule for a profile in Control D.
Update Custom Rule by Rule IDTool to update an existing custom rule by its ID.
Move Profile Rule to FolderTool to move a specific custom rule into a different folder.
Update Profile ScheduleTool to update a specific schedule within a profile.
Modify Service for ProfileTool to modify a specific service rule for a profile.

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

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: ["control_d"],
  });

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

  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 control_d, 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 Control d 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 Control d MCP Agent with another framework

OpenAI Agents SDK logo

OpenAI Agents SDK

Use Control d MCP with OpenAI Agents SDK

Claude Agent SDK logo

Claude Agent SDK

Use Control d MCP with Claude Agent SDK

Claude Code logo

Claude Code

Use Control d MCP with Claude Code

Claude Cowork logo

Claude Cowork

Use Control d MCP with Claude Cowork

Codex logo

Codex

Use Control d MCP with Codex

OpenClaw logo

OpenClaw

Use Control d MCP with OpenClaw

Hermes logo

Hermes

Use Control d MCP with Hermes

CLI logo

CLI

Use Control d MCP with CLI

Google ADK logo

Google ADK

Use Control d MCP with Google ADK

LangChain logo

LangChain

Use Control d MCP with LangChain

Mastra AI logo

Mastra AI

Use Control d MCP with Mastra AI

LlamaIndex logo

LlamaIndex

Use Control d MCP with LlamaIndex

CrewAI logo

CrewAI

Use Control d MCP with CrewAI

FAQ

What are the differences in Tool Router MCP and Control d MCP?

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

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

Yes, absolutely. You can configure which Control d 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 Control d 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.

Get started for freeGet a demo↗