How to integrate Brex 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

Brex logo
Vercel AI SDK logo
divider

Introduction

This guide walks you through connecting Brex to Vercel AI SDK v6 using the Composio tool router. By the end, you'll have a working Brex agent that can create a new card for our intern, get the latest account statement pdf, add an expense for last week's conference through natural language commands.

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

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

Also integrate Brex with

TL;DR

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

The Brex MCP server is an implementation of the Model Context Protocol that connects your AI agent and assistants like Claude, Cursor, etc directly to your Brex account. It provides structured and secure access to your company's financial tools, so your agent can perform actions like managing cards, tracking expenses, creating budgets, and retrieving account details on your behalf.

  • Card creation and management: Instantly issue new cards, assign them to users or vendors, and control spending limits via your agent.
  • Automated expense tracking: Let your agent create and log expenses, helping you keep company finances organized and up to date.
  • Budget and department setup: Quickly create budgets for teams or projects and set up new departments without manual effort.
  • Account and statement retrieval: Fetch detailed account information and download statements whenever you need financial insights or records.
  • User and location management: Add new users or locations to your Brex account, making it easy to scale as your business grows.

Supported Tools & Triggers

Tools
Add Webhook Group MembersAdd webhook subscription members to a webhook group.
Archive BudgetArchive a budget to mark it as inactive and prevent future expenses.
Archive Spend LimitArchive a spend limit in Brex.
Create BudgetCreate a new budget for departments or projects.
Create Spend Limit (Budget V1)Tool to create a new Spend Limit (Budget) in Brex.
Create CardCreate a new card and assign it to a user.
Create DepartmentCreate a new department.
Create Document Upload for ReferralCreate a document upload URL for a referral and return a pre-signed S3 upload URL.
Create ExpenseCreate a new expense.
Create Custom FieldCreate a new custom field in Brex.
Create Field ValuesCreate custom field values for a specific Brex field.
Create LocationCreate a new location.
Create Referral RequestTool to create a new referral to Brex for onboarding.
Create Spend LimitTool to create a spend limit in Brex.
Create Job TitleCreate a new job title.
Create UserCreate a new user in the Brex account.
Create VendorCreate a new vendor in Brex for payment operations.
Create Webhook GroupTool to create a webhook group for targeting webhook subscriptions.
Create Webhook SubscriptionRegister a new webhook subscription to receive real-time notifications for Brex events.
Delete FieldTool to delete a custom field by Brex ID.
Delete Field ValuesDelete custom field values for a specific field.
Delete VendorDelete a vendor by ID.
Get Account StatementsGet finalized statements for the primary cash account.
Get Budget DetailsGet detailed information about a specific budget.
Get Card DetailsGet detailed information about a specific card.
Get Card ExpenseTool to retrieve detailed information about a specific card expense by its ID.
Get Card NumberGet card number, CVV, and expiration date for a specific card.
Get Card TransactionsGet settled card transactions for the primary card account.
Get Company Cash AccountsRetrieve all cash (deposit) accounts for the company.
Get Company DetailsGet company information associated with the OAuth2 access token.
Get Department By IDGet detailed information about a specific department by ID.
Get ExpenseTool to get details of a specific expense by ID.
Get Field By IDTool to retrieve a custom field by its Brex ID.
Get Field Value By IDTool to retrieve a specific field value by field ID and field value ID.
Get Legal EntityTool to retrieve a legal entity by its ID from Brex.
Get Location By IDGet detailed information about a specific location by ID.
Get Referral By IDTool to retrieve a specific referral by its ID from Brex.
Get Spend Limit By IDTool to get detailed information about a specific spend limit by its ID.
Get Spend LimitsGet spend limits configuration.
Get Title by IDGet detailed information about a specific job title by ID.
Get Transaction by IDGet details of a specific transaction by ID.
Get Card TransactionsGet card transactions from the primary Brex account.
Get Transactions by Amount RangeGet card transactions filtered by amount range and date period.
Search Transactions by DescriptionSearch and filter card transactions by description text.
Get User LimitRetrieves the monthly spending limit and available balance for a specific Brex user.
Get User ProfileGet user profile information from Brex Team API.
Get Vendor By IDTool to get vendor details by ID.
Get Webhook GroupTool to retrieve details of a specific webhook group by ID.
Get Webhook SubscriptionTool to retrieve details of a specific webhook subscription by ID.
List Budget ProgramsTool to list all budget programs in the organization.
List BudgetsList all budgets and show available amounts across all cards.
List Card AccountsTool to list all card accounts for the company.
List CardsList all cards associated with the account.
List Card StatementsTool to list finalized statements for primary card accounts.
List DepartmentsList all departments in the organization.
List ExpensesTool to list expenses from the Brex platform.
List Field ValuesTool to list values under a custom field.
List Legal EntitiesTool to list all legal entities in the Brex account with pagination support.
List LocationsList all locations in the organization.
List ReferralsTool to list all referrals created in the Brex account.
List Job TitlesList all job titles in the organization.
List TransfersLists transfers for the account.
List TripsLists trips according to the filters passed in the query string.
List UsersList all users in the Brex account.
List VendorsList all vendors.
List Webhook Group MembersTool to list all members of a webhook group.
List Webhook GroupsTool to list all webhook groups.
List Webhook SecretsTool to retrieve webhook signing secrets for validating incoming webhook messages from Brex.
List Webhook SubscriptionsTool to list all registered webhook subscriptions.
Create Receipt MatchCreates a receipt match request and returns a pre-signed S3 upload URI.
Create Receipt Upload for ExpenseCreates a receipt upload request for a specific card expense and returns a pre-signed S3 upload URL.
Set User LimitSet or update the monthly spending limit for a Brex user.
Transfer CardTransfer a card to a different user.
Update BudgetUpdate an existing budget's configuration in Brex.
Update Spend Limit (V1)Tool to update a Spend Limit (budget) using the v1 API.
Update Card LimitsUpdate spending limits and controls for a Brex card.
Update Card StatusUpdate the status of a Brex card by locking, unlocking, or terminating it.
Update ExpenseTool to update an expense by its ID.
Update FieldTool to update a custom field in Brex.
Update Field ValuesTool to update custom field values in Brex for a specific field.
Update Spend LimitUpdates an existing Brex spend limit by its ID.
Update UserUpdate user details.
Update VendorTool to update vendor information in Brex.
Update Webhook SubscriptionUpdates an existing Brex webhook subscription's configuration.

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: ["brex"],
  });

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

  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 brex, 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 Brex 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 Brex MCP Agent with another framework

OpenAI Agents SDK logo

OpenAI Agents SDK

Use Brex MCP with OpenAI Agents SDK

Claude Agent SDK logo

Claude Agent SDK

Use Brex MCP with Claude Agent SDK

Claude Code logo

Claude Code

Use Brex MCP with Claude Code

Claude Cowork logo

Claude Cowork

Use Brex MCP with Claude Cowork

Codex logo

Codex

Use Brex MCP with Codex

OpenClaw logo

OpenClaw

Use Brex MCP with OpenClaw

Hermes logo

Hermes

Use Brex MCP with Hermes

CLI logo

CLI

Use Brex MCP with CLI

Google ADK logo

Google ADK

Use Brex MCP with Google ADK

LangChain logo

LangChain

Use Brex MCP with LangChain

Mastra AI logo

Mastra AI

Use Brex MCP with Mastra AI

LlamaIndex logo

LlamaIndex

Use Brex MCP with LlamaIndex

CrewAI logo

CrewAI

Use Brex MCP with CrewAI

FAQ

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

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

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

Yes, absolutely. You can configure which Brex 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 Brex 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↗