VTEX Checkout MCP. Simulate Carts and Order Flows from Chat.
Gemini Works with every AI agent you already use
…and any MCP-compatible client
Just plug in your AI agents and start using Vinkius.
VTEX Checkout MCP Server connects your VTEX e-commerce checkout API to any AI agent. It lets you simulate full order flows, calculate accurate shipping costs, apply discount coupons, and manage client profiles—all from natural conversation.
Stop clicking through admin dashboards; get pre-purchase data instantly.
What your AI agents can do
Add coupon
Applies a specific coupon code to an existing shopping cart, showing the resulting discount and new total.
Create address
Registers a completely new shipping address record for a customer profile.
Get client profile
Retrieves the primary details and contact information for an existing client using their ID number.
The agent calculates final totals, shipping rates, and potential discounts for any set of items and postal code.
You fetch the complete state of a shopping cart, including line items, client ID, and payment conditions.
The agent pulls specific client data—name, CPF/CNPJ, email—by referencing a user's unique ID.
You test coupon codes against an active cart and see the precise discount applied to the total cost.
The agent allows you to register new, valid shipping addresses for a client profile.
Ask AI about this MCP
Supported MCP Clients
Gemini Waiting for input…
VTEX Checkout MCP Server: 6 Tools for E-commerce Ops
These tools let your AI client interact directly with the core functions of a VTEX store, enabling complex tasks like simulating orders or managing client data.
019d761dadd coupon
Applies a specific coupon code to an existing shopping cart, showing the resulting discount and new total.
019d761dcreate address
Registers a completely new shipping address record for a customer profile.
019d761dget client profile
Retrieves the primary details and contact information for an existing client using their ID number.
019d761dget orderform
Fetches all current data associated with a specific shopping cart, including items, discounts, and payment terms.
019d761dsimulate order
Calculates the final cost of a potential order by combining product totals and available shipping rates for a given postal code.
019d761dsimulate payment
Validates payment tokens against simulated gateway rules to confirm if a transaction would process successfully before checkout.
Choose How to Get Started
Build a custom MCP for your own tools, or connect a ready-made integration from our catalog.
Build Your Own
Turn any API into an MCP. Import a spec, define Agent Skills, or deploy with MCPFusion.
- Import from OpenAPI, Swagger, or YAML specs
- Create Agent Skills with progressive disclosure
- Deploy to edge with MCPFusion framework
- Built in DLP, auth, and compliance on every call
- Real time usage dashboard and cost metering
- Publish to catalog or keep private
Make Your AI Do More
Start with VTEX Checkout, then connect any of our 4,700+ other servers whenever your AI needs more. One click, no limits.
- Use this MCP plus 4,700+ others, all in one place
- Add new capabilities to your AI anytime you want
- Every connection is secured and compliant automatically
- Track usage and costs across all your servers
- Works with Claude, ChatGPT, Cursor, and more
- New servers added to the catalog every week
What you can do with this MCP connector
Look, you gotta connect your VTEX e-commerce checkout API to an AI agent using this server. It handles everything that goes down between adding some goods to a cart and actually hitting 'Pay Now.' You'll be able to run complete order simulations, check client details, and test discounts without ever having to log into the damn admin panel.
This lets your AI client manage full shopping flows just by chatting with it.
Simulating Full Order Transactions
You can run a deep dive on any potential purchase using simulate_order. You feed it items, quantities, and a specific postal code, and it spits out the final cost, factoring in available shipping rates. This doesn't book anything; it just calculates what the total should be before checkout.
Managing Cart State and Discounts
If you need to know exactly what's sitting in an active cart, you use get_orderform. That tool pulls up every piece of data associated with that shopping cart: all line items, any existing client IDs attached, and the payment terms. If you find a coupon code, you run it through add_coupon.
This instantly applies the specific discount to the current total, letting you see the exact savings amount and the new final price point.
Handling Client Logistics
You don't have customer details? You can use get_client_profile with a client's ID number; that pulls up their full contact info—name, CPF/CNPJ, email. If they need somewhere to ship stuff and you don't have it on file, run create_address. That registers a brand-new shipping address record right into the customer profile.
Validating Payment and Shipping
Before your agent lets someone pay, you gotta make sure the payment token works. You run simulate_payment to validate payment tokens against simulated gateway rules; it confirms if a transaction would process successfully without actually taking any money. Finally, when calculating totals, remember that shipping rates are critical—the initial simulation handles those rates based on the provided postal code.
How VTEX Checkout MCP Works
- 1 First, subscribe and provide your VTEX Account Name, App Key, and App Token. These credentials link the server to your store's backend.
- 2 Next, you instruct your AI agent (Claude, Cursor, etc.) to perform a task—for example: 'Check the total for 3 units of Product X shipping to ZIP Y.'
- 3 The agent maps that request to the appropriate tools (
simulate_order,add_coupon, etc.), calls the server, and returns a structured result with accurate totals and options.
The bottom line is: it takes complex API sequences—like checking inventory, calculating taxes, and applying coupons—and turns them into simple function calls your agent can execute in plain language.
Who Is VTEX Checkout MCP For?
E-commerce Managers who spend hours manually testing promotions across different regions. Support Engineers who need to look up a customer's cart status mid-conversation without leaving the chat window. Operations Analysts who validate complex shipping rules for new product launches.
They use get_client_profile and get_orderform instantly during a live chat to tell the customer exactly what's in their cart or who they are, without needing to transfer them to another team.
They run simulations using simulate_order and add_coupon when testing flash sales or new promotions. They need instant feedback on totals before launching anything live.
They validate the entire checkout flow by chaining tools: first, checking client details (get_client_profile), then simulating payment (simulate_payment) to confirm regional rules work.
What Changes When You Connect
- Stop manual lookups. You can instantly get the full cart status by calling
get_orderform, giving support agents real-time answers without leaving their chat interface. - Test promotions before launch. Use
add_couponcombined withsimulate_orderto confirm discount accuracy and final pricing for any product mix and location. - Accelerate customer onboarding. If a client needs an address added, use
create_addressdirectly via your agent instead of making them navigate a separate form. - Pre-validate the purchase path. Running
simulate_paymentconfirms if payment conditions will pass before you hit the actual checkout button, saving time and errors. - Deep visibility into client data. Quickly pull name, ID, and contact info using
get_client_profile, eliminating the need to search multiple database tabs.
Real-World Use Cases
A customer asks: 'How much will this cost with shipping?'
The support agent doesn't guess. The agent uses simulate_order, passing the items and the customer’s ZIP code. The response immediately lists standard, express, and free shipping rates along with the final calculated total.
A user has an old cart and needs a discount.
The agent first runs get_orderform to confirm the items are still in the cart. Then, it calls add_coupon using the customer-provided code. The system reports the exact discount amount and the new total instantly.
A client is calling from a new location.
Instead of asking the customer to fill out a form, the agent uses create_address with the provided data. This registers the address in the system while keeping the conversation moving forward.
Testing a complex sale before a big launch.
The E-commerce Manager chains tools: they call get_client_profile first, then use add_coupon, and finally run simulate_order. This sequence validates client eligibility, discount application, and shipping costs all in one go.
The Tradeoffs
Calculating totals manually
A support agent tries to figure out the total by adding up item prices and then guessing taxes/shipping rates based on memory. This is slow and often wrong.
→
Always run simulate_order. Pass it the full cart state (retrieved via get_orderform) and the shipping ZIP code. The tool handles all the complex calculation logic for you.
Assuming payment works
The agent tells the user, 'Your card will work.' This is dangerous because it doesn't account for regional rules or specific token failures.
→
Before confirming a purchase path, use simulate_payment. This validates the token and payment conditions against VTEX’s internal logic without actually charging the customer.
Missing cart context
Trying to run add_coupon or simulate_order without a specific orderform ID. The system fails because it doesn't know which cart to apply changes to.
→
Always start by running get_orderform if you need the current context, ensuring your subsequent calls have the correct active Order Form ID.
When It Fits, When It Doesn't
Use this server if your primary goal is pre-purchase validation or customer service data retrieval. You need to know what a total will look like—including discounts and shipping—before it actually happens. The combination of get_orderform, add_coupon, and simulate_order makes the perfect workflow for this.
Don't use this if you need actual, live payment processing involving irreversible charges or bank-level fraud checks. This server focuses on simulation and data retrieval. For real-time financial transactions that require immediate settlement, you must connect to a dedicated gateway service instead of relying solely on simulate_payment.
Independent Platform Disclaimer: Vinkius is an independent platform and is not affiliated with, endorsed by, sponsored by, verified by, or otherwise authorized by VTEX. 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.
VINKIUS INFRASTRUCTURE
Cloud Hosted
Managed infra
V8 Isolated
Sandboxed per request
Zero-Trust Proxy
No stored credentials
DLP Enforced
Policy on every call
GDPR Compliant
EU data residency
Token Compression
~60% cost reduction
Works with Claude, ChatGPT, Cursor, and more
The Model Context Protocol standardizes how applications expose capabilities to LLMs. Instead of operating in isolation, your AI gains direct access to external platforms, live data, and real-world actions through secure, standardized connections.
This server provides 6 capabilities that interface natively with Claude, ChatGPT, Cursor, and any MCP client. No middleware. No custom integration required.
Available Capabilities
Checking out used to mean endless clicks across multiple systems.
Today, if you're supporting an order or testing a sale, you run into friction. You open the admin panel, search for the customer ID, copy their cart contents, then open a separate tab just to check shipping rates by ZIP code. If you want to test a coupon, you often have to manually adjust line items and wait for the system to recalculate everything.
With this MCP server, all those steps collapse into one conversation. Your agent calls `get_client_profile` for identity, runs `get_orderform` for contents, then uses `add_coupon` and `simulate_order`—all in sequence. You get the final total instantly, without leaving the chat.
VTEX Checkout MCP Server: Run full checkout ops from your agent.
You eliminate switching between customer databases, inventory checkers, and pricing calculators. The server handles the complexity of knowing if an address is valid (`create_address`), how a discount interacts with tax rates (`add_coupon`), and what the final shipping cost will be.
It moves checkout logic from being a series of backend clicks into a simple conversation flow. Your AI client simply asks, 'What's the total for this?' and gets an accurate answer.
Common Questions About VTEX Checkout MCP
How do I get my VTEX API credentials? +
Log in to your VTEX Admin, go to Account Settings → Application Keys, and create a new App Key. You'll receive an App Key and App Token pair. You also need your Account Name (the subdomain you use to access the VTEX Admin, e.g., 'mystore'). No code, no SDK — just connect and go.
Can my agent simulate a full cart with shipping costs? +
Yes! Use the 'simulate_order' tool with a list of items and a postal code. Your agent will return the full cart breakdown — totals, per-item pricing, available logistics options, estimated delivery dates, and applicable discounts — all without touching the actual store.
What happens when a customer asks support about their current cart? +
Your support agent can use the 'get_orderform' tool to instantly retrieve the customer's active cart — items, quantities, prices, applied coupons, and shipping address. No need to switch tabs or navigate the admin. The agent can then apply a coupon or check shipping costs, all within the same conversation.
Can I validate payment conditions before placing bulk orders? +
Absolutely. The 'simulate_payment' tool lets your agent validate payment tokens and check accepted conditions — perfect for operations teams running bulk order validations across multiple sellers and payment methods without processing real transactions.
If I call the `get_orderform` tool, what happens if the shopping cart ID is invalid or has expired? +
The system returns a structured error code indicating the cart's status. This tells your agent exactly why the data failed to load—whether it’s due to an incorrect orderform ID or because the session timed out.
When I use `get_client_profile`, what unique identifiers do I need to provide for accurate lookups? +
You must provide a primary identifier, such as the user's CPF/CNPJ or email address. Using multiple pieces of identifying data increases confidence and helps the agent find the correct profile quickly.
Does calling `add_coupon` require that I first run `create_address` for the client? +
No, you can apply coupons independently. However, if your goal is to simulate a final purchase total, it's best practice to create an address before running simulate_order.
What are the rate limits or performance considerations when I run multiple cart simulations with `simulate_order`? +
The service is designed for high throughput. For extremely rapid, bulk simulation tasks, monitor the response time and ensure your agent handles potential throttling signals gracefully.
Use it with your favorite AI tools
Connect this server to Cursor, Claude, VS Code, and more.
More in this category
LeadDyno
Empower your AI to process LeadDyno affiliate links, manage program members, and track generated commission sales instantly.
Rebrickable LEGO
Explore the LEGO universe — search sets, parts, minifigs, themes and colors with full catalog data from any AI agent.
Ticketmaster
Search for live events, venues, and attractions, and manage ticket reservations directly through Ticketmaster's APIs.
You might also like
ExerciseDB
Access 1300+ exercises with body parts, target muscles, equipment, instructions and animated GIFs.
Douyin Open Platform
Orchestrate Douyin (TikTok China) content — manage videos, handle comments, and track user analytics directly from any AI agent.
Mailchimp
Synchronize marketing contacts — list audiences, control subscriber limits, and edit campaign vectors autonomously.