Commerce.js MCP. Manage your entire store checkout flow via 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.
Commerce.js manages your entire online store workflow through a single API connection. Use this server in any AI client to list products, build carts, process checkouts, and track orders without logging into a dashboard.
It gives your agent direct access to product catalog metadata, real-time cart updates, and secure order capture.
What your AI agents can do
Add to cart
Adds a specified product to an existing shopping cart.
Capture order
Finalizes and captures a customer order using payment details, completing the purchase.
Create cart
Initializes or retrieves an existing shopping cart session for a user.
Retrieves metadata for specific items or lists full product catalogs, including filtering by category.
Creates new shopping carts and allows the agent to add multiple products or adjust quantities on existing line items.
Generates a temporary checkout token from a populated cart, setting up the necessary data for payment processing.
Captures an order using customer details and the generated tokens to complete the transaction in the backend system.
Lists recent orders or fetches full order records, allowing agents to track fulfillment status without manual login.
Browses and lists all available product categories, helping map the store's internal hierarchy.
Ask AI about this MCP
Supported MCP Clients
Gemini Waiting for input…
Commerce.js: 10 Tools for E-commerce Operations
These tools allow your agent to perform the core functions of an online store, handling everything from product discovery and cart building through secure checkout completion.
019ea5e5add to cart
Adds a specified product to an existing shopping cart.
019ea5e5capture order
Finalizes and captures a customer order using payment details, completing the purchase.
019ea5e5create cart
Initializes or retrieves an existing shopping cart session for a user.
019ea5e5generate checkout token
Creates a secure, temporary token needed to process payment from the current cart items.
019ea5e5get order
Retrieves all details for one specific order ID (requires Secret Key).
019ea5e5get product
Fetches detailed metadata and current pricing for a single product by its ID.
019ea5e5list categories
Returns a list of all primary categories defined in the store's product structure.
019ea5e5list orders
Retrieves a paginated list of recent order IDs and summary data (requires Secret Key).
019ea5e5list products
Lists the store's product catalog, allowing filtering by category or pagination.
019ea5e5update cart item
Changes the quantity of a specific product already placed in the cart.
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 Commerce.js, 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
Commerce.js MCP Server - E-commerce API Tools
Listen up. This server gives your agent direct, hands-on control over every part of your online store's backend process. You don't have to log into a dashboard or mess with manual forms; you just tell the AI client what you need done, and it uses these tools to make it happen instantly.
It handles everything from listing product details all the way through capturing a final payment.
Product Lookup & Store Navigation:
You wanna know what's for sale? You can list every single item in your catalog using list_products, and you can filter that whole mess by specific categories or handle pagination if there are too many results. If you need the deep dive on a single item, get_product fetches all the metadata and current pricing just by knowing the product ID.
You also don't have to guess what sections your store has; running list_categories gives you every primary category defined in your structure.
Cart Manipulation:
When it comes to shopping carts, you gotta start somewhere. Use create_cart to initialize a new session or pull up an existing cart for the user. Once the cart exists, your agent can add products using add_to_cart, specifying which product ID and how many units they want. If the customer changes their mind about the quantity of something already in the cart, you adjust it with update_cart_item.
It's all done programmatically.
Secure Checkout Preparation:
Before money moves, you gotta prepare the payment data. The process starts by taking all the items currently sitting in the user’s shopping cart and using generate_checkout_token to create a secure, temporary token. This token holds all the necessary data—the product IDs, quantities, and prices—ready for the final step.
Order Finalization & Tracking:
To complete the sale, your agent uses the customer details and that generated token in capture_order. That function actually sends the signal to finalize the purchase and captures the funds in the backend system. If you need visibility into what's happening after the sale, you can use list_orders to grab a paginated rundown of recent order IDs and summary data (you'll need your Secret Key for this).
To get the full picture on just one specific transaction, run get_order, feeding it only the unique order ID. This gives you all the details—the status, the items purchased, everything—without needing manual login access.
How Commerce.js MCP Works
- 1 Subscribe to this server and enter your specific e-commerce Public Key and Secret Key credentials.
- 2 Your AI client uses those keys to authenticate with the Commerce.js API endpoints.
- 3 The agent executes a tool call (e.g.,
list_products) which returns live data directly into the chat conversation.
The bottom line is, your AI client talks shop using your store's actual backend tools; no manual logins needed.
Who Is Commerce.js MCP For?
This is for anyone who spends time managing e-commerce operations but hates the dashboard. Think support reps who are tired of copy-pasting order IDs, or developers needing to test checkout logic without a local sandbox environment. If your job involves checking inventory status or processing returns, this server saves clicks.
Needs to check specific customer order details (get_order) instantly to answer 'Where is my stuff?' questions without transferring the call to fulfillment.
Quickly checks current inventory levels or lists recent orders (list_orders) to verify stock before a marketing push, all from a simple chat command.
Tests complex checkout logic by running full sequences—create_cart, add_to_cart, generate_checkout_token, and capture_order—directly from their coding environment.
What Changes When You Connect
- You get real-time inventory checks. Instead of guessing if a product is in stock, the agent runs
get_productto fetch current data before you recommend it. - Order tracking moves instantly from manual lookup to automation. Use
get_orderorlist_ordersto pull up fulfillment status for any order ID without touching the dashboard. - Cart management is faster than clicking through pages. You can use
create_cartand then quickly adjust items withupdate_cart_item—all in a few lines of conversation. - Checkout preparation happens securely, step-by-step. The agent uses
generate_checkout_tokento get the necessary payment parameters before it even attempts tocapture_order. - Product discovery is streamlined. Running
list_productslets you see your catalog structure and filter items by category in one go.
Real-World Use Cases
Customer asks about a past order
A customer texts, 'Can I check status for ord_987?' Your agent doesn't ask them to log in. It immediately calls get_order using the ID and pulls back fulfillment details instantly, providing a seamless answer.
Developer needs to test checkout edge cases
A developer wants to simulate adding multiple items. They run create_cart, then use add_to_cart for Item A and Item B, followed by generate_checkout_token. This simulates a full user journey without live payment risks.
Support team needs catalog depth
A support agent says, 'What are your main product lines?' The agent calls list_categories to map the store hierarchy. They then use this data to call list_products for specific groups.
Manager needs to update a pending sale
A manager notices an order cart has too much stock of one item. Instead of calling support, they run create_cart, use add_to_cart, and then correct the amount using update_cart_item before generating the token.
The Tradeoffs
Treating the API like a simple search box
Asking the agent simply, 'Show me my products.' The AI might just return a list of links or names without structured data needed for calculation.
→
You must first run list_products to get the full catalog and then use get_product if you need detailed specs (like weight or dimensions) for a specific item ID.
Skipping security steps
Trying to capture an order immediately just by providing names, assuming payment details are stored. This will fail because the API requires preparation.
→
Always run create_cart first. Then, use add_to_cart to populate it, and finally, call generate_checkout_token. Only after that sequence can you confidently attempt capture_order.
Confusing listing with retrieval
Trying to get details for a specific order by just naming the customer. The system needs a unique identifier, not general context.
→
Use list_orders first to find the correct Order ID. Once you have that unique ID, pass it into the dedicated get_order tool.
When It Fits, When It Doesn't
Use this server if your primary bottleneck is converting e-commerce data—products, carts, and orders—into actionable chat commands. You need to automate tasks like 'Check stock for X' or 'Finalize order Y.' Don't use it if you just want to browse the store; that's a frontend job. Also, don't rely on this solely for payment processing; while generate_checkout_token and capture_order handle the data flow, actual payment gateway setup is external. If your main goal is inventory reporting without transactional capability, simply using list_products might be enough, but if you need to act (add, update, capture), this suite of tools is mandatory.
Independent Platform Disclaimer: Vinkius is an independent platform and is not affiliated with, endorsed by, sponsored by, verified by, or otherwise authorized by Commerce.js. 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 10 capabilities that interface natively with Claude, ChatGPT, Cursor, and any MCP client. No middleware. No custom integration required.
Available Capabilities
Checking order status shouldn't require three different tabs and a password reset.
Right now, checking an order means logging into the dashboard, finding the customer’s name or ID, navigating to 'Order History,' clicking the specific order number, and then manually reviewing the fulfillment status. It's slow, and it requires multiple credentials.
With this MCP server, you simply ask your agent: 'What is the status of ord_987?' The agent executes `get_order` directly against the API and gives you a clean answer immediately. You don't touch a password or click a tab.
Commerce.js MCP Server lets you manage cart flow with `create_cart`, `add_to_cart`, and `update_cart_item`.
Without this server, adjusting a shopping cart—like changing the quantity of an item or removing it—means logging in, finding the specific product ID, navigating to the edit page, manually calculating the new subtotal, and saving. It's tedious manual work.
Now, you just tell your agent: 'I need 3 units of prod_Q123.' The tool handles the cart initialization (`create_cart`), adds the correct quantity (`add_to_cart`), and updates the total, giving you accurate, real-time results instantly.
Common Questions About Commerce.js MCP
How do I use Commerce.js with `list_products` to find an item? +
Run list_products first; this gives you the catalog and IDs. If you want more details on a specific product, grab its ID and run get_product. It pulls down all the metadata for that single item.
What's the sequence for capturing an order using `capture_order`? +
You must first initialize the cart with create_cart, then populate it with items using add_to_cart. Only after that do you generate a secure token via generate_checkout_token before calling capture_order.
Can I check order status without knowing the Order ID? (Using `list_orders`) +
No. You must use list_orders to pull a list of recent orders and find the correct Order ID first. Then, you pass that unique ID into get_order to get full details.
Does `update_cart_item` work if I haven't created a cart yet? +
No. You must run create_cart first. The system needs an existing cart session ID before it can modify quantities or line items using update_cart_item.
How do I authenticate my AI client to use any of the Commerce.js tools? +
You must provide your Chec Public Key and Secret Key during setup. The agent needs these credentials first. Once connected, you can call all available methods like create_cart or list_products directly from your preferred client.
What information is required when I use the `generate_checkout_token` tool? +
It requires a valid Cart ID and customer details. The token generation step needs to know which cart to finalize and who the order belongs to. This prepares the data structure for actual payment processing.
How do I use `get_product` to check specific product details or inventory? +
Just pass the Product ID directly into the tool call. It returns detailed metadata, including current stock levels and pricing specifics. This is faster than listing all products just to find one item's status.
If I need to understand my store structure before searching for items, should I use `list_categories`? +
Yes, running list_categories shows your entire product hierarchy. This maps out the available paths in your store's catalog. You can then filter subsequent calls using these category IDs.
Use it with your favorite AI tools
Connect this server to Cursor, Claude, VS Code, and more.
More in this category
Drip
Manage subscribers, campaigns, and events in Drip via your AI Agent.
Airbnb
Search and manage Airbnb listings, experiences, reservations, and pricing directly from any AI agent.
Redo
Manage returns, track shipping protection claims, and oversee exchanges via AI agents with Redo.
You might also like
NOAA Space Weather — Solar & Geomagnetic Intelligence
Real-time space weather intelligence: planetary Kp geomagnetic index, 3-day Kp forecast, solar wind speed and magnetic field, aurora probability forecast (Ovation model), solar flux (F10.7), and Dst storm index from NOAA SWPC.
Copysmith
Generate marketing copy, product descriptions, and ad variations at scale with AI trained on high-performing content.
HubSpot CRM (Full)
Manage contacts, companies, deals, tickets, notes, owners, and pipelines — full access to your HubSpot CRM through natural conversation.