How to Use the Marqeta MCP in OpenAI Agents SDK

Q: How do I connect the Marqeta MCP Server to the OpenAI Agents SDK?

Install openai-agents. Initialize MCPServerStreamableHttp with your Vinkius endpoint and pass that directly into the mcpservers list when you build your agent.

Q: Can my OpenAI agent provision cards for Apple Pay via Marqeta?

Yes. The agent can call provisionapplepay or provisiongooglepay on an active card token. You just need to ensure the card is in an active state first.

Q: How does Marqeta handle webhooks with the OpenAI Agents SDK?

Your agent can configure endpoints using createwebhook and test them with pingwebhook. The OpenAI tracing dashboard will show you exactly what payload the agent sent during setup.

Q: Does the OpenAI Agents SDK cache Marqeta tool lists?

You should set cacheToolsList=True when initializing the server. This prevents the agent from re-fetching the 31 available endpoints every time it boots.

Connect Marqeta to the OpenAI Agents SDK to issue cards and trace payment flows directly in your production agent system.

See Vinkius in Action

Works with every AI agent you already use

…and any MCP-compatible client

MCP Servers - Free for Subscribers

Connect Marqeta MCP to OpenAI Agents SDK

Create your Vinkius account to connect Marqeta to OpenAI Agents SDK and route execution through our secure gateway. The platform manages server hosting, runtime updates, and security layers. Configuration requires no manual server provisioning.

GDPR Free for Subscribers

Setup Marqeta with OpenAI Agents SDK

Ask AI about this MCP

ChatGPT

Claude

Perplexity

Issue Cards with OpenAI Agents SDK

Your agent needs to issue cards, but you don't want it randomly spinning up thousands of accounts. Using `create_user` and `create_card`, you can script exact provisioning flows. The SDK's built-in guardrails validate the agent's logic before it touches the Marqeta API. You can set up specialized agents. One handles identity by running `perform_kyc` and `create_business`. If that passes, it hands off to a provisioning agent that calls `create_card_product` and `provision_apple_pay`. Everything gets logged in your OpenAI dashboard.

Simulate Authorizations and Clearings

Testing payment logic usually requires waiting on external processors. This MCP server lets your agent mock the entire transaction lifecycle. It can trigger `simulate_authorization` to check if a card has funds, then immediately run `simulate_clearing` to settle it. If something goes wrong, the agent can call `simulate_reversal`. Because the OpenAI Agents SDK traces every step, you can see exactly which simulated webhook failed and why the transaction was declined.

Manage Velocity and Funding Controls

Unrestricted spending breaks corporate card programs. Your agent can enforce limits dynamically by calling `create_velocity_control` and `create_auth_control`. You define the rules, and the agent applies them to specific card tokens. For funding, the agent manages the ledger. It checks `get_balances` and triggers `create_gpa_order` or `create_program_funding_source` when accounts run low. The SDK ensures the agent doesn't loop and over-fund the account.

Setup guide

Set up Marqeta MCP in OpenAI Agents SDK

Prerequisites

Python 3.10+ installed
openai-agents package (pip install openai-agents)
Active Vinkius subscription with a valid endpoint token

1

Install the SDK

Run pip install openai-agents to install the OpenAI Agents SDK. The MCP integration is built-in — no extra dependencies needed.
2

Connect via SSE transport

Use MCPServerSse with your Vinkius endpoint URL. Replace [YOUR_TOKEN_HERE] with your token from cloud.vinkius.com. The SDK auto-discovers all Marqeta tools at runtime.
3

Create your Agent

Pass the MCP to Agent(mcp_servers=[server]). The agent receives Marqeta tools as native definitions — JSON schemas resolve automatically.
4

Run the agent

Call Runner.run(agent, prompt) to execute. The agent invokes the appropriate Marqeta tools and returns structured results. Copy the full example on the right to get started.

agent.py

import asyncio
from agents import Agent, Runner
from agents.mcp import MCPServerSse

async def main():
    async with MCPServerSse(
        url="https://edge.vinkius.com/[YOUR_TOKEN_HERE]/mcp"
    ) as server:
        agent = Agent(
            name="Marqeta Agent",
            instructions="You have access to Marqeta tools.",
            mcp_servers=[server],
        )
        result = await Runner.run(agent, "List recent transactions")
        print(result.final_output)

asyncio.run(main())

Get your connection token →

Independent Platform Disclaimer: Vinkius is an independent platform and is not affiliated with, endorsed by, sponsored by, verified by, or otherwise authorized by Marqeta. All third-party trademarks, logos, and brand names are the property of their respective owners. Their use on this website is strictly for informational purposes to identify service compatibility and interoperability.

Why Choose Vinkius

Vinkius connects your tools to AI with real-time monitoring and automatic cost savings — all from one dashboard.

Connect Marqeta now

Real-time monitoring

Live

visibility into every interaction

Connect your favorite tools to your AI and see exactly what's happening — every request, every response, in real time.

Built-in savings

60%

lower AI costs

Vinkius compresses data between your apps and your AI automatically. Lower bills every month — no configuration required.

Single dashboard

One

place for every integration

Every tool your AI connects to, managed from a single screen. One account, complete control.

Common questions about Marqeta MCP in OpenAI Agents SDK

Install `openai-agents`. Initialize `MCPServerStreamableHttp` with your Vinkius endpoint and pass that directly into the `mcp_servers` list when you build your agent.

Yes. The agent can call `provision_apple_pay` or `provision_google_pay` on an active card token. You just need to ensure the card is in an active state first.

Your agent can configure endpoints using `create_webhook` and test them with `ping_webhook`. The OpenAI tracing dashboard will show you exactly what payload the agent sent during setup.

You should set `cacheToolsList=True` when initializing the server. This prevents the agent from re-fetching the 31 available endpoints every time it boots.

This integration handles highly sensitive PII, KYC data, and raw card balances. The MCP architecture keeps your API keys isolated in the Vinkius sandbox, meaning the agent only receives temporary tokens and authorization statuses, never the underlying root credentials.

Use it with your favorite AI tools

Connect this server to Cursor, Claude, VS Code, and more.

OpenAI Agents SDK sdk-python

Google ADK sdk-python

Pydantic AI sdk-python

Vercel AI SDK sdk-typescript