FairBite skill

FairBite plans a McDonald's China order for 2-6 people within budget, tastes and exclusions, prices checked by the official MCP.

  • 120 ★ on the repository
  • Last commit
  • MIT
Skill

Extended access

Why this level

  • Uses an official McDonald's MCP token for menu and price queries, and its output may be wrong about dish composition.
All reasons and checks

lzp76995-labb/mcd-fairbite

Install

In your terminal, with SkillFoxx CLI

npx skillfoxx add skills/fairbite

Detects the agents on your machine, shows the access level and pins the version.

Other ways to install

Access and checks

How we review

Extended accessThe tool runs code or makes network calls and only reads external data.

  • Uses an official McDonald's MCP token for menu and price queries, and its output may be wrong about dish composition.

This is third-party code. Review the repository files before installing.

What it does

FairBite splits a group order for 2-6 people so it fits both the shared budget and each person's limit. Dishes are filtered by exclusions first, then the lowest taste score is raised, then the overall score, and price is cut last. The skill reads the menu and calculates prices through the official MCP `https://mcp.mcd.cn` and does not create or pay for orders. A web demo runs with `npm start` at `http://127.0.0.1:4173`, with no dependencies beyond Node.js 22.

Who it is for. People organizing a team lunch in China who are tired of reconciling budgets and tastes in chat, and developers exploring the official McDonald's MCP.

Good fit when

  • Four people order lunch with a shared budget of 150 yuan and a 40 limit each.
  • One person avoids pork and another prefers chicken, so each needs a separate pick.
  • You want to compare strategies: best for the least satisfied person versus lowest total.
  • You want to verify the total with the official price calculation before ordering in the app.

Not a fit when

  • You need decisions for severe allergies: composition data is incomplete, so confirm with the restaurant.
  • The order must be placed and paid automatically: the skill does not do that.
  • You are outside China or have no McDonald's MCP token.

Example request

There are four of us, a shared budget of 150 yuan and at most 40 each: find a store near a busy square, plan the order, then verify the price with the official calculation.

Limitations

The project is independent and not a McDonald's product. The algorithm searches a limited list of 1-12 candidates and does not promise the lowest price across the whole menu. Taste scores come from rules, not real preferences. Prices and availability depend on the official MCP query at run time. Calls need your own `MCD_MCP_TOKEN`, so never publish it. Verification covers only the documented test at one Shanghai store, and the author did not check allergies, delivery or coupons. The project was made for a developer contest.

How to disable. Delete the `mcd-fairbite` folder from your skills directory, remove the McDonald's MCP entry from the client config and revoke the token in the open.mcd.cn console.

FAQ

Does the demo need a model key?

No. The web demo starts with `npm start` on Node.js 22 with no third-party dependencies and no model API key.

Does the skill place the order?

No. The documented live run created no orders, coupons or payments, only time, menu, detail and price calculation queries.

How do I check the output?

`npm test` runs planner and MCP tests. Compare the total with the official server's `calculate-price` response.

How many people are supported?

From 2 to 6 people, 1 to 12 candidates and a budget up to 2000 yuan.

How do I run tests?

Run `npm test`. For the scenario demo use `npm run demo`, and `node scripts/plan.mjs examples/demo.json --json` prints a plan.

Discussion

No questions or reviews yet.

Skills

McButler works out calories and the full cost of an order, including travel and queue time, from the McDonald's China MCP

SkillExtended access208Repository stars

A skill on the McDonald's China MCP: turns a budget and wishes into up to three menus with official prices, leftover budget and coupon terms

SkillExtended access1.3KRepository stars

McDonald's China ordering skill: picks a restaurant, asks party size and budget, offers three sets with price checks

SkillBroad access123Repository stars

CLI and skill for McDonald's China orders: pick a meal by calories, price it with coupons and hold the order until confirmed.

CLIBroad access129Repository stars

A skill for ordering from McDonald's China via its official MCP: coupon-aware prices, nutrition, order status and a quiet summary.

SkillBroad access118Repository stars

Comparison with similar tools

Cards that share the main search query, and how they differ.

ToolOfficialRepository starsMain difference
FairBite (this card)no120
mcd-party-plannerno1,382plans an order for a group; FairBite computes the budget and compares fairness strategies
mcd-budget-mealno1,327picks a budget meal for one person; this one splits a shared limit and personal limits across several people

README in short

The planner works in integer fen, and recommendations can be copied as a split list or downloaded as JSON. The live run is documented with a date as a historical price snapshot.

README badge

Report a problem