How to integrate Paystack MCP with Vercel AI SDK v6

Framework Integration Gradient
Paystack Logo
Vercel AI SDK Logo
divider

Introduction

This guide walks you through connecting Paystack to Vercel AI SDK v6 using the Composio tool router. By the end, you'll have a working Paystack agent that can list all successful transactions from this week, create a new customer profile in paystack, refund a recent payment from customer through natural language commands.

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

The Paystack MCP server is an implementation of the Model Context Protocol that connects your AI agent and assistants like Claude, Cursor, etc directly to your Paystack account. It provides structured and secure access so your agent can perform Paystack operations on your behalf.

Supported Tools & Triggers

Tools
Add Products to PageTool to add products to a Paystack payment page.
Add Products to StorefrontTool to add previously created products to a Storefront.
Add Split Code to Virtual TerminalTool to add a split code to a virtual terminal.
Add Subaccount to SplitTool to add a subaccount to a split configuration, or update the share of an existing subaccount.
Archive Payment RequestTool to archive a payment request to clean up records.
Assign Destination to Virtual TerminalTool to assign WhatsApp destinations to a Virtual Terminal for payment notifications.
Check Pending ChargeTool to check the status of a pending charge transaction.
Check Slug AvailabilityTool to check if a custom slug is available for use when creating a payment page.
Bulk Create Transfer RecipientsCreate multiple transfer recipients in batches.
Create ChargeTool to initiate a payment by integrating the payment channel of your choice.
Create CustomerTool to create a customer on your Paystack integration.
Create Payment PageTool to create a payment page on your Paystack integration.
Create Payment RequestCreate a new payment request by issuing an invoice to a customer.
Create PlanTool to create a plan for recurring payments on Paystack.
Create RefundTool to initiate a refund for a previously completed transaction on Paystack.
Create Split ConfigurationTool to create a split configuration for transaction revenue sharing.
Create StorefrontCreate a digital storefront on Paystack to manage and display products.
Create SubaccountTool to create a subaccount on your Paystack integration.
Create Transfer RecipientTool to create a new recipient for transfers.
Create Virtual TerminalCreate a Virtual Terminal on your Paystack integration to accept payments.
Deactivate AuthorizationTool to deactivate an authorization when a card or payment method needs to be forgotten.
Deactivate Virtual TerminalTool to deactivate a Virtual Terminal on your integration.
Delete ProductDelete a previously created product from Paystack by its ID.
Delete StorefrontDelete a previously created storefront from Paystack by its ID.
Delete Transfer RecipientTool to delete a transfer recipient by setting it to inactive status.
Disable OTP for TransfersTool to disable OTP requirement for programmatic transfers.
Disable SubscriptionTool to disable a subscription on your Paystack integration.
Duplicate StorefrontTool to duplicate a previously created Storefront.
Enable OTP for TransfersTool to enable OTP requirement for transfers.
Export TransactionsTool to export transactions from Paystack for a specific timeframe.
Fetch BalanceTool to fetch the available balance on your Paystack integration.
Fetch balance ledgerTool to fetch all pay-ins and pay-outs that went into your Paystack balance.
Fetch Bulk Charge BatchTool to retrieve a specific bulk charge batch by code.
Fetch CustomerTool to get details of a customer on your integration.
Fetch Mandate AuthorizationsTool to fetch the list of direct debit mandate authorizations associated with a customer.
Fetch Payment PageTool to fetch details of a payment page on your Paystack integration.
Fetch Payment Request DetailsTool to fetch details of a payment request from Paystack.
Fetch Payment Session TimeoutTool to fetch the payment session timeout configured on your Paystack integration.
Fetch PlanTool to get details of a plan on your integration.
Fetch ProductTool to get details of a product on your integration.
Fetch Product OrdersTool to fetch all orders for a particular product on your Paystack integration.
Fetch RefundTool to get details of a refund on your integration.
Fetch SplitTool to get details of a split configuration for a transaction.
Fetch StorefrontTool to get the details of a previously created Storefront.
Fetch Storefront OrdersTool to fetch all orders in a Paystack storefront.
Fetch SubaccountTool to get details of a subaccount on your integration.
Fetch SubscriptionTool to get details of a subscription on your integration.
Fetch TransactionTool to get details of a transaction carried out on your integration.
Fetch Transaction TimelineTool to fetch the timeline of a transaction showing the steps taken from initiation to completion.
Fetch Transfer RecipientTool to fetch details of a transfer recipient.
Fetch Virtual TerminalTool to fetch a Virtual Terminal on your Paystack integration.
Finalize Payment RequestTool to finalize the creation of a draft payment request for a customer.
Generate Update Subscription LinkTool to generate a link for updating the card on a subscription.
Get Payment Request TotalsTool to retrieve metrics for all pending and successful payment requests.
Get Transaction TotalsTool to get the total amount received on your Paystack account.
Initialize AuthorizationTool to initialize a reusable authorization for recurring transactions.
Initialize TransactionTool to initialize a transaction from your backend.
Initiate Bulk ChargeTool to initiate a bulk charge on your integration.
List BanksTool to get a list of all supported banks and their properties on Paystack.
List Bulk Charge BatchesTool to list bulk charge batches created on your integration.
List Charges in BatchTool to retrieve the charges associated with a specified bulk charge batch.
List countriesTool to retrieve a list of countries that Paystack currently supports.
List CustomersTool to list customers available on your integration.
List DisputesTool to list transaction disputes filed by customers on your Paystack integration.
List Apple Pay DomainsTool to list all registered Apple Pay domains on your integration.
List Mandate AuthorizationsTool to get a list of all direct debit mandates on your integration.
List OrdersTool to list orders available on your integration.
List Payment PagesTool to list payment pages available on your Paystack integration.
List Payment RequestsTool to list all previously created payment requests to your customers.
List PlansTool to list all subscription plans available on your integration.
List ProductsTool to list products available on your integration.
List RefundsTool to list refunds available on your integration.
List SettlementsTool to list settlements made to your settlement accounts.
List SplitsTool to list the transaction splits available on your integration.
List states for address verificationTool to retrieve a list of states for a country for address verification.
List Storefront ProductsTool to list all products in a Paystack storefront.
List StorefrontsTool to list storefronts available on your integration.
List SubaccountsTool to list subaccounts available on your integration.
List SubscriptionsTool to list all subscriptions available on your integration.
List TransactionsTool to list transactions that have occurred on your integration.
List Transfer RecipientsTool to list transfer recipients available on your integration.
List TransfersTool to list all transfers carried out on your integration.
List Virtual TerminalsTool to list virtual terminals available on your integration.
Pause Bulk ChargeTool to pause processing of a bulk charge batch.
Publish StorefrontTool to make a Storefront publicly available on Paystack.
Remove Subaccount from SplitTool to remove a subaccount from a split configuration.
Resend OTP for TransferTool to resend OTP for a transfer.
Resolve Card BINTool to get basic information about a card BIN (Bank Identification Number).
Resume Bulk Charge BatchTool to resume processing of a previously paused bulk charge batch.
Send Payment Request NotificationTool to send an email notification to a customer for a previously created payment request.
Send Subscription Update LinkTool to send an email to the customer with a link to update the card on their subscription.
Set Risk ActionTool to set a customer's risk action by whitelisting or blacklisting them.
Submit PhoneTool to submit phone number when requested during a charge transaction.
Trigger direct debit activation chargeTool to trigger an activation charge on an inactive direct debit mandate for a customer.
Unassign Destination from Virtual TerminalTool to unassign WhatsApp destinations from a Virtual Terminal.
Unregister Apple Pay DomainTool to unregister a top-level domain or subdomain previously used for Apple Pay integration.
Update CustomerTool to update a customer's details on your Paystack integration.
Update Payment PageTool to update a payment page on your Paystack integration.
Update Payment RequestTool to update a payment request on your Paystack integration.
Update Payment Session TimeoutTool to update the payment session timeout on your Paystack integration.
Update PlanTool to update a plan's details on your Paystack integration.
Update ProductTool to update a product on your integration.
Update Split ConfigurationTool to update a split configuration for transactions.
Update StorefrontTool to update the details of a previously created Storefront.
Update SubaccountTool to update a subaccount's details on your Paystack integration.
Update Transfer RecipientTool to update the details of a transfer recipient.
Validate Bank AccountTool to validate the authenticity of a customer's bank account before sending money.
Validate CustomerTool to validate a customer's identity on your Paystack integration.
Validate OrderTool to validate a pay for me order on Paystack.
Verify Payment RequestTool to verify the status of a previously created payment request on Paystack.
Verify Storefront SlugTool to verify the availability of a slug before using it for a Storefront.
Verify TransactionTool to confirm the status of a transaction using its reference.

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

What is Tool Router?

Composio's Tool Router 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 Tool Router

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

How the Tool Router works

The Tool Router 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: ["paystack"],
  });

  const mcpUrl = session.mcp.url;
What's happening:
  • We're creating a Tool Router session that gives your agent access to Paystack 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 Paystack-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 Paystack 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 paystack, 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("\nšŸ‘‹ 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 Paystack 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 Paystack 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: ["paystack"],
  });

  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 paystack, 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("\nšŸ‘‹ Session ended.");
    process.exit(0);
  });
}

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

Conclusion

You've successfully built a Paystack 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 Paystack MCP Agent with another framework

FAQ

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

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

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

Yes, absolutely. You can configure which Paystack 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 Paystack 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.