Meituan Waimai MCP for AI. Manage order status, refunds, and menus from conversation.
Gemini Works with every AI agent you already use
…and any MCP-compatible client
Connect to your AI in seconds.
Meituan Waimai connects your restaurant operations to any AI client. Use this server to manage the entire food delivery lifecycle: check order status, confirm new orders, process refunds, and update menu stock—all via conversation.
It’s essential for multi-location operators who need real-time control over their shop floor.
What your AI can do
Cancel order
Cancels a specific Meituan delivery order, requiring you to provide a mandatory reason for the cancellation.
Complete order
Marks an order as finished, signaling the customer has received the food and triggering payment settlement.
Confirm order
Changes an order from pending status to confirmed, starting the preparation process for the restaurant team.
You can list orders by status (pending, confirmed, etc.), confirm new ones, or mark them as delivered using get_order_list, confirm_order, and mark_delivering.
Use handle_refund to review customer complaints and either approve or reject refund claims after checking order history.
Access the menu via list_menus, check current stock levels, and update availability using update_stock so customers can't order what you don't have.
Get deep details on a single order—including customer info and items—by calling get_order_detail with the correct order ID.
Check operational data like business hours or delivery coverage for any specific location using get_restaurant_info (POI ID).
Ask an AI about this
Waiting for input…
Meituan Waimai MCP Server: 10 Tools for Order & Menu Control
These tools let your AI client manage every stage of restaurant operations, from order listing and confirmation to inventory adjustments and refund processing.
Make your AI actually useful.
Add this MCP to Claude, Cursor, or Windsurf and your AI stops guessing. It gets real tools to look things up, take action, and handle the stuff you keep doing by hand.
Start using Meituan Waimai on VinkiusCancel Order
Cancels a specific Meituan delivery order, requiring you to provide a mandatory reason for the cancellation.
Complete Order
Marks an order as finished, signaling the customer has received the food and...
Confirm Order
Changes an order from pending status to confirmed, starting the preparation process...
Get Order Detail
Retrieves all specific information about one order ID, including contents, address...
Get Order List
Lists multiple orders for a location, allowing filtering by status like pending or...
Get Restaurant Info
Gets general business data—like hours or coverage area—for any specific restaurant POI ID.
Handle Refund
Processes a refund request, letting you approve or reject the claim and providing a mandatory reason for your decision.
List Menus
Pulls the entire menu catalog for a restaurant, showing item names, prices, and...
Mark Delivering
Changes an order's status to 'out for delivery,' optionally adding rider tracking...
Update Stock
Adjusts the quantity of a menu item, marking it as sold out or restocking inventory...
Security and governance baked right in.
Pick your AI client below to get set up. Just create a Vinkius account, subscribe, and you're instantly up and running. We handle the entire backend infrastructure, delivering out-of-the-box support for HTTPS Streamable, SSE, and OAuth2—zero messy routing required.
Claude AI
Open Claude Settings
Go to claude.ai, click your profile icon, then navigate to Customize → Connectors.
Add Custom Connector
Click the "+" button and select Add custom connector. Paste your Vinkius endpoint URL:
https://edge.vinkius.com/[YOUR_TOKEN_HERE]/mcp
Replace [YOUR_TOKEN_HERE] with your token
from cloud.vinkius.com. For OAuth-protected servers, expand
Advanced settings to add credentials.
Start a conversation
Open a new chat. The Meituan Waimai integration is available immediately — no restart needed.
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 Meituan Waimai, then connect any of our 5,100+ other servers whenever your AI needs more. One click, no limits.
- Use this MCP plus 5,100+ 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
Independent Platform Disclaimer: Vinkius is an independent platform and is not affiliated with, endorsed by, sponsored by, verified by, or otherwise authorized by Meituan Waimai. 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 connection provides 10 powerful capabilities that interface natively with Claude, ChatGPT, Cursor, and other compatible AI platforms. No middleware. No custom integration required.
The current system forces you into endless dashboard clicking.
Right now, handling daily orders means logging into the merchant portal, navigating to the 'Order List' tab, filtering by 'Pending,' then opening each order ID one by one. If there are 30 pending orders, that’s 30 clicks just to review them. Then, if you need to check stock for a specific item, you have to open another menu tab and search.
With this MCP Server, you simply tell your agent: 'Show me all pending orders and verify the stock of appetizers.' The agent executes `get_order_list` and then runs `list_menus`. You get both results compiled in one chat window. It’s immediate, it's structured, and it requires zero clicks from you.
Meituan Waimai MCP Server: Order & Menu Control
Before this, updating stock or processing a refund was a multi-step process that required specific forms and multiple confirmations. If you missed one step—like failing to provide the cancellation reason for `cancel_order`—the whole transaction would fail, costing time and money.
Now, the agent handles the sequence. You give the high-level command: 'Process refund for ORD-789 because of late delivery.' The agent automatically runs `get_order_detail`, verifies the claim, and then executes `handle_refund` in one go. It’s about executing complex workflows through simple conversation.
What your AI can actually do with this
Meituan Waimai connects your restaurant ops straight into your AI client. Instead of logging into some bloated merchant dashboard, you manage everything—from confirming an order to adjusting how much stock you have on hand—all through conversation. This server is built for multi-location operators who need real-time control over the shop floor without leaving their chat window.
When you run into complex, multi-step tasks that used to require manual clicks and jumping between different systems, this tool gives your agent the ability to handle it all in sequence. You'll be able to manage the entire food delivery lifecycle through conversation alone.
Managing Order Flow
You can get a bird’s eye view of what’s happening with get_order_list, which lets you pull up multiple orders for your location and filter them by status, whether they're pending or already delivered. When a new order comes in, use confirm_order to change its status from pending to confirmed; this kicks off the prep work for your kitchen team.
If an order gets stuck or needs attention, you can grab all the specifics on a single job using get_order_detail, which pulls up everything like customer notes and item contents based on the correct order ID. You've got options if things go south; you can use cancel_order to shut down a specific delivery, but you gotta provide a mandatory reason why.
Once the food is out the door, you mark it as delivering using mark_delivering, and you can even optionally add rider tracking details so the customer knows exactly where their food is. When everything is done, use complete_order to mark the order as finished—that signals the customer got the food and triggers payment settlement.
Handling Money and Disputes
When customers complain or something goes wrong financially, you manage it all with handle_refund. This tool lets you review a refund claim and either approve it or reject it; in either case, you gotta provide a mandatory reason for your decision so the records are clean.
Running Inventory and Menu Control
You don't have to guess what sells. You can pull your entire menu catalog using list_menus, which shows item names, current prices, and—most importantly—your real-time stock levels. If you run low on something, use update_stock right away; this adjusts the quantity of a specific menu item, letting you mark it as sold out or restock inventory instantly so customers can't order what you don't have.
You'll also find that when you need general info about your business—like operational hours or delivery coverage for a location—you use get_restaurant_info, just by providing the specific POI ID.
This gives you total control, letting your agent handle the whole operation without needing to switch between tabs.
019d8457-17f0-7040-bbfc-a9cbc67ec98b 
Here's how it actually works
The bottom line is, your AI client runs complex, multi-step business processes using simple chat commands.
Subscribe to the Meituan Waimai server and provide your App ID/App Secret credentials.
Your AI client uses a conversational prompt, telling the agent what needs doing (e.g., 'I need to confirm all pending orders for POI123').
The agent executes the necessary tool calls (get_order_list, then confirm_order), and returns the final status or required action to you.
Who is this actually for?
This tool is for restaurant owners and operations managers who hate logging into multiple vendor dashboards. If you spend time cross-referencing order statuses with inventory sheets, this saves hours. It’s built for multi-location teams that need a single interface to run the whole operation.
Monitoring incoming orders and approving refunds without having to log into the merchant portal.
Querying order volumes across multiple POIs (Points of Interest) and coordinating status changes for staff.
Quickly checking get_order_detail for new orders or marking items as sold out using update_stock when ingredients run low.
Investigating refund claims by running handle_refund and pulling full order history via get_order_list.
What Changes When You Connect
Streamlined Order Flow: Use get_order_list to pull all pending orders instantly. Instead of manually checking the dashboard for new requests, your agent gathers them in one list so you can confirm everything at once.
Real-Time Inventory Control: Don't let customers order what you sold out of. By calling update_stock, you immediately change item availability on the customer menu, preventing lost sales and frustration.
Faster Dispute Resolution: When a refund claim comes in, use get_order_detail to pull all necessary data (items, price, address) in seconds. Then, run handle_refund with confidence, knowing your decision is backed by verifiable order history.
Single Source of Truth: Need to check if the restaurant is open or if delivery covers a certain area? Run get_restaurant_info using only the POI ID—no need to Google it or call the store directly. It keeps all location data centralized.
Automated Fulfillment Tracking: When an order moves from confirmed to out for delivery, use mark_delivering. This updates the customer tracking status automatically, improving transparency without manual intervention.
See it in action
The Morning Rush: Handling 50 Pending Orders
A manager wakes up and sees a flood of pending orders. Instead of clicking through the dashboard fifty times to confirm each one, they ask their agent: 'Confirm all pending orders for POI123.' The agent runs get_order_list and then executes bulk confirmations using confirm_order. Done in seconds.
The Inventory Crisis
A key ingredient runs out mid-day. Instead of waiting for a customer to complain or the system to fail, the kitchen staff member tells their agent: 'Mark chicken as sold out.' The agent executes update_stock, and the item instantly disappears from the customer menu.
The Refund Investigation
A customer claims they never received their food. Instead of asking the customer for more proof, the service agent runs get_order_detail using the order ID. They check the delivery status and use that data to run handle_refund, providing a clear, documented reason for approval or denial.
Multi-Location Coordination
The Ops Manager needs to know if three different store locations (POI A, POI B, POI C) are even open today. They run get_restaurant_info for all three IDs in one command, getting instant status checks and business hours.
The honest tradeoffs
Relying on Screenshots
Taking screenshots of order lists or menus to show a manager later. This requires manually cropping images and cross-referencing dates.
Use get_order_list and get_order_detail. The agent pulls the raw, verifiable JSON data directly into your chat history, eliminating image ambiguity.
Manual Status Updates
A delivery rider calling in to manually update a status or confirming an order by hitting 'OK' on multiple screens. This is slow and prone to human error.
Use mark_delivering or confirm_order. These tools handle the state transition automatically, updating both your system record and the customer view.
Ignoring Stock Dependencies
Accepting an order for a popular item without checking if you actually have enough ingredients. This leads to backorders and complaints.
Always run list_menus first to check current stock levels, then use that data before allowing any confirmations or accepting new orders.
When It Fits, When It Doesn't
Use this server if your workflow involves managing the complete lifecycle of an order—from initial listing and confirmation through delivery tracking, financial settlement (refunds), and menu maintenance. You need a system that treats operations as conversational tasks.
Don't use it if all you need is simple read-only data about one specific field (e.g., just the restaurant name). For isolated reads, general API wrappers might suffice. However, if your task involves changing state—like canceling an order (cancel_order), confirming a new order (confirm_order), or updating stock (update_stock)—this MCP Server is necessary because it ties those actions together in a single conversational flow.
If you only need to query data from a different platform (e.g., accounting software), then use that specific service's API wrapper instead.
Questions you might have
How do I use the `confirm_order` tool? +
You must provide both the order ID and the restaurant POI ID. This tool moves an order from a waiting state to confirmed, starting the preparation process for kitchen staff.
What is the difference between `get_order_list` and `get_order_detail`? +
get_order_list gives you a summary view of many orders (status, total price). get_order_detail drills down into one specific order to see every item, customer note, or address.
Can I use `handle_refund` without providing a reason? +
No. The API requires you to provide a clear explanation for the refund decision (both approval and rejection). This ensures your decision is documented correctly.
`update_stock` needs a food ID—where do I get it? +
You must first run list_menus. That tool provides the full catalog, which includes the necessary Food IDs you need to pass into the update_stock function.
Before I run any order tool, how do I verify my restaurant's location using `get_restaurant_info`? +
Use this tool with your Point of Interest (POI) ID. It confirms the restaurant’s existence, checks current business hours, and validates delivery coverage for that specific location before you try to manage orders.
What are the critical conditions I must know when using `cancel_order`? +
You absolutely must provide a detailed reason; canceling negatively impacts merchant ratings. Also, the order has to be in a state that allows cancellation—it can't already be marked as delivered or completed.
When exactly should I call `complete_order`? +
Only run this tool after the customer physically receives the food. Calling it too early will fail and won't trigger the payment settlement to your merchant account.
How do I connect my system and get started with the Meituan Waimai API? +
You need to register through the developer portal to generate your specific App ID and App Secret. These credentials authenticate your agent and allow it to interact with your restaurant data.
How do I get my Meituan Waimai App ID and App Secret? +
Log in to the Meituan Waimai Developer Console, navigate to API Documentation > Access Guide, and register your application. You'll receive an App ID and App Secret. Make sure your application has the required API permissions for order management, menu access, and refund handling. The signature algorithm uses MD5 hashing of sorted parameters combined with your App Secret.
What's the order lifecycle in Meituan Waimai? +
The typical order flow is: 1) Pending (待确认, status=1) — Customer placed order, awaiting merchant confirmation. 2) Confirmed (已确认, status=3) — Merchant accepted the order, kitchen starts preparation. 3) Delivering (配送中, status=5) — Food is out for delivery with rider. 4) Completed (已完成, status=7) — Customer received the order, payment settled. 5) Cancelled (已取消, status=8) — Order was cancelled by merchant or customer. At each stage, you can query details, and merchants can confirm, mark delivering, or complete the order.
How do I handle stock management and sold-out items? +
Use the update_stock tool to set stock quantities for menu items. Set stock=0 to mark an item as sold out (it will immediately disappear from the customer menu). Set a positive number to replenish stock. Get the food ID from the list_menus tool first. This is critical during peak hours when ingredients run out — quickly updating stock prevents customers from ordering unavailable items and reduces refund requests.
We've already built the connector for Meituan Waimai. Just plug in your AI agents and start using Vinkius.
No hosting. No infrastructure. No complex setup.
All 10 tools are live and waiting.
You're up and running in seconds.
Gemini Vinkius gives your AI agents access to the full catalog of app connectors, all fully managed, secure, and enterprise-ready. One subscription, every tool you need.
Built, hosted, and secured by Vinkius. You just connect and go.