Find the cheapest McDonald's China order with coupons
Developer hujjun's WorkBuddy skill gathers McDonald's China menus and coupons to find the cheapest order; the author reports a ¥47 Big Mac meal dropping to ¥21.90 in sample data.
Done with WorkBuddy
What they did
Developer hujjun built a skill for the WorkBuddy agent for a McDonald's developer contest, using a connector to McDonald's China's ordering system. When a user asks for the cheapest way to order, the agent gathers stores, menu, prices, nutrition and coupons from four sources. A fixed Python program then works out the cheapest valid combination. The price is rechecked and the user confirms before the order is created.
How it went
On sample data, the author reports a Big Mac, medium fries and Coke drop from ¥47.00 to ¥21.90, a 53% saving, using a 1+1 combo coupon. They note the sample prices and coupons are demo snapshots, so real prices depend on the store.
Worth knowing
Testing with a real account, the author found one coupon source misses the best deal: some coupons must be claimed first, some are store-only, and some work only for pickup. The math is left to code because AI models often get prices and discounts wrong.
Try it yourself with WorkBuddy
I want to order [items] from [restaurant] near [your location], for [pickup/delivery], with a budget of [amount] and up to [calorie limit]. Use the restaurant's ordering tools to fetch the menu, prices and every available coupon, including ones I can claim but haven't yet. Do the price math in code, not in your head. Respect each coupon's stacking rules and channel limits, and include delivery fees. Show me the top 3 cheapest valid combinations with their totals and the coupons used. Re-check the final price, and wait for my confirmation before you place any order or start payment.
Swap the highlighted [parts] for your own details.
Source: GitHub · hujjun · Oct 9, 2026
Written with AI assistance from the original post and checked against our editorial guidelines. Results are as reported by the source, not independently verified.
