How to integrate Route4me MCP with Vercel AI SDK v6

Trusted by
AWS
Glean
Zoom
Airtable

30 min · no commitment · see it on your stack

Route4me logo
Vercel AI SDK logo
divider

Introduction

This guide walks you through connecting Route4me to Vercel AI SDK v6 using the Composio tool router. By the end, you'll have a working Route4me agent that can list all available delivery vehicles today, mark stop 12 on route 8 as completed, show vehicles with capacity over 1000 lbs through natural language commands.

This guide will help you understand how to give your Vercel AI SDK agent real control over a Route4me account through Composio's Route4me MCP server.

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

Also integrate Route4me with

TL;DR

Here's what you'll learn:
  • How to set up and configure a Vercel AI SDK agent with Route4me integration
  • Using Composio's Tool Router to dynamically load and access Route4me 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 Route4me MCP server, and what's possible with it?

The Route4me MCP server is an implementation of the Model Context Protocol that connects your AI agent and assistants like Claude, Cursor, etc directly to your Route4me account. It provides structured and secure access to your fleet and route data, so your agent can perform actions like listing vehicles, retrieving asset details, updating destination statuses, and streamlining last-mile logistics on your behalf.

  • Full vehicle inventory retrieval: Instantly ask your agent to pull a complete list of all vehicles in your organization for planning and operational audits.
  • Vehicle data access for route planning: Get detailed information on each vehicle—including capacity and assignment details—to optimize delivery routes.
  • Automated destination status updates: Let your agent mark destinations as completed, skipped, or failed after deliveries or stops, keeping your records accurate in real time.
  • Streamlined delivery confirmation workflows: Have your agent confirm delivery outcomes automatically, ensuring your route progress is always up to date for dispatchers and drivers.

Supported Tools & Triggers

Tools
Add Order Custom FieldTool to add a new custom field definition for orders.
Add Route NoteTool to add a note to a route or address destination.
Create AddressTool to add a new address to the address book.
Create Avoidance ZoneTool to create an avoidance zone with circular, rectangular, or polygon shape.
Create Location TypeTool to create a new location type in the address book.
Create Master RouteTool to create a new master route in Route4Me.
Create Custom Note TypeTool to create a new custom note type in Route4me.
Create Route OptimizationTool to create a new route optimization in Route4Me.
Create Optimization ProfileTool to create a new optimization profile in Route4Me.
Create Order Custom FieldTool to create a new custom field definition for orders.
Create Orders from ContactsTool to create orders from address book contacts.
Create TerritoryTool to create a new territory with circular, rectangular, or polygon shape.
Create Vehicle Capacity ProfileTool to create a vehicle capacity profile with weight, volume, item count, and revenue limits.
Delete Address from RouteTool to delete an address from a route.
Delete AddressesTool to delete addresses from the Route4Me address book.
Delete Avoidance ZoneTool to delete an avoidance zone by territory_id.
Delete Configuration SettingTool to delete a member configuration setting by config_key.
Delete Current UserTool to delete the current authenticated user account.
Delete Location TypeTool to delete a location type from the address book.
Delete OptimizationTool to remove optimization problems from the Route4Me database.
Delete Optimization ProfileTool to delete an optimization profile by ID.
Delete Order Custom FieldTool to delete an order custom field by ID.
Delete Route V4Tool to delete one or more routes by route_id.
Delete TerritoryTool to delete a territory by territory_id.
Delete Vehicle Capacity Profile by IDTool to delete a vehicle capacity profile by ID.
Dispatch RoutesTool to approve or revoke routes for execution in Route4Me.
Display Datatable ConfigTool to retrieve datatable configuration for vehicle capacity profiles.
Duplicate RoutesTool to duplicate one or more existing routes.
Find AddressTool to retrieve detailed information about a specific address from the Route4Me address book.
Geocode AddressTool to geocode an address and obtain its coordinates.
Geocode Address (Body)Tool to geocode an address by sending a body payload with the address and options.
Bulk Geocode AddressesTool to geocode multiple addresses in a single request.
Geocode Address ForwardTool to convert an address to geographic coordinates using forward geocoding.
Get Activity FeedTool to retrieve activities from the Route4Me activity feed with optional filtering and pagination.
Get AddressTool to get address/destination information from a route.
Get Address Book GroupsTool to retrieve address book groups from Route4Me.
Get Addresses from Address BookTool to retrieve addresses from the Route4Me address book.
Get Async Export Tracking HistoryTool to save tracking history file via async export.
Get All Territories (Avoidance Zones v5)Tool to retrieve all avoidance zones (territories) using API v5.
Get Avoidance Zone By IDTool to retrieve a specific avoidance zone by its territory ID.
Get Clustered LocationsTool to retrieve locations grouped by geographic proximity using geohash clustering.
Get Current Member InformationTool to retrieve the current authenticated member's information.
Get Activity FeedTool to retrieve all activities from the Route4Me activity feed.
Get Last Scheduled Route for MemberTool to retrieve the most recently scheduled route for a specific member as of today.
Get Location HeatmapTool to retrieve a heat map of location clusters from the address book.
Get Location Service-Time HeatmapTool to retrieve location service-time heatmap data for address book locations.
Get Location TypeTool to get location type details by ID.
Get Member Configuration SettingsTool to get member configuration settings with optional filtering.
Get Note Custom TypesTool to retrieve all custom note types configured for the account.
Get Optimization ProfileTool to retrieve detailed information about an optimization profile by its unique identifier.
Get Route Destination By IDTool to get a route address/destination by its ID.
Get Route InfoTool to get detailed route information by route ID.
Get Route ListTool to retrieve a list of routes with filtering and pagination.
Get Route SettingsTool to get route settings with detailed information including waypoints, optimization settings, and delivery details.
Get Territories with Location CountsTool to retrieve territories with location counts for geographic analysis.
Get TerritoryTool to retrieve territory information by ID or get all territories.
Submit Device Tracking DataTool to submit device GPS tracking information to Route4Me.
Get Tracking History ExportTool to export tracking history file from Route4Me.
Get User By IDTool to retrieve a sub-user by their unique ID.
Get User LocationsTool to view locations of all users/drivers in real-time.
Get User Modification HistoryTool to get user modification history from Route4Me.
Get User TemplateTool to retrieve the template of a sub-user with detailed information including contact information, preferences, and account status.
Get User Information (v4)Tool to retrieve authenticated user information in v4 format.
Get VehiclesTool to retrieve details of vehicles in the organization.
Get Vehicle Capacity Profile As DatatableTool to retrieve vehicle capacity profiles in datatable format with pagination.
Get Vehicle Capacity Profile by IDTool to retrieve a specific vehicle capacity profile by its ID.
Get Vehicle Capacity Profile Page ConfigTool to retrieve vehicle capacity profile page configuration including unit settings.
Get Vehicle Capacity ProfilesTool to retrieve vehicle capacity profiles with weight, volume, and item constraints.
List Location TypesTool to retrieve location type definitions used in the address book system.
List Optimization ProfilesTool to retrieve optimization profiles with settings and configuration details.
List OptimizationsTool to get a list of all optimizations or retrieve a single optimization by ID.
List Order Custom FieldsTool to retrieve all order custom field definitions from Route4Me v5.
List Route DirectionsTool to get detailed route directions including turn-by-turn navigation information for each stop.
List Team UsersTool to get the list of team members/users in the account.
List Users By IDsTool to retrieve detailed information about team members by their member IDs.
List Team Users (Paginated)Tool to get a paginated list of team members/users in the account.
Lookup For New DestinationTool to lookup optimal routes for inserting a new destination.
Merge RoutesTool to merge multiple routes into a single target route.
Move Route DestinationTool to move a destination from one route to another.
Preview Route AssignmentTool to generate a preview of route assignments based on filters, grouping, sorting, and assignment strategy without applying changes.
Re-run OptimizationTool to re-optimize an existing optimization problem by updating route parameters and/or addresses.
Reverse Geocode AddressTool to reverse geocode coordinates to obtain an address.
Set Destination StatusTool to set or update the status of a route destination.
Set Route RatingTool to set a rating for a route.
Set Route Tracking DataTool to insert tracking data for a route.
Update AddressTool to update an existing address in the Route4Me address book.
Update Avoidance ZoneTool to update an existing avoidance zone.
Update Location TypeTool to update an existing location type in the address book.
Update Member Configuration SettingTool to update a member configuration setting.
Update Custom Note TypeTool to update an existing custom note type in Route4Me.
Update Optimization ProfileTool to update an existing optimization profile.
Update Order Custom FieldTool to update an existing order custom field in Route4Me.
Update RoutesTool to bulk update multiple routes in a single operation.
Update Route SettingsTool to update route settings including optimization parameters, distance units, and device types.
Update Route V4Tool to update route parameters and addresses in Route4Me.
Update TerritoryTool to update an existing territory.
Update Vehicle Capacity Profile By IDTool to update an existing vehicle capacity profile by ID.

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: ["route4me"],
  });

  const mcpUrl = session.mcp.url;
What's happening:
  • We're creating a Tool Router session that gives your agent access to Route4me 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 Route4me-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 Route4me 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 route4me, 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 Route4me 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 Route4me 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: ["route4me"],
  });

  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 route4me, 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 Route4me 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 Route4me MCP Agent with another framework

FAQ

What are the differences in Tool Router MCP and Route4me MCP?

With a standalone Route4me MCP server, the agents and LLMs can only access a fixed set of Route4me tools tied to that server. However, with the Composio Tool Router, agents can dynamically load tools from Route4me 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 Route4me tools.

Can I manage the permissions and scopes for Route4me while using Tool Router?

Yes, absolutely. You can configure which Route4me 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 Route4me 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.