SkillFoxx

MCD Fit Planner skill

An unofficial skill for McDonald's China: a training plan, combo comparison by price and macros, order tracking and recaps

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

Broad access

Why this level

  • The skill creates real McDonald's China orders over MCP after confirmation, with payment in the official channel
  • The official MCP token opens menus, prices and orders on your account
All reasons and checks

tomsbin/mcd-fit-plannerAuthor:tomsbin

Install

In your terminal, with SkillFoxx CLI

npx skillfoxx add skills/mcd-fit-planner

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

Other ways to install

Access and checks

How we review

Broad accessThe tool can change data, run commands or handle secrets.

  • The skill creates real McDonald's China orders over MCP after confirmation, with payment in the official channel
  • The official MCP token opens menus, prices and orders on your account

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

What it does

MCD Fit Planner sits on top of the official McDonald's China MCP and helps choose food around a training plan. The user enters once the training days, high-carb days and rest days, daily and per-meal targets and a budget. Then on a request like "training day, keep my burger, find a cheaper combo" the skill finds combos in the chosen store's menu, checks their real components, queries the final price and shows calories, protein, fat and carbs. If an item could be swapped for a cheaper or closer-to-target one, the skill shows the difference in price and nutrition and waits for confirmation. An order is created only after confirmation and a repricing, and the user pays through the official channel. What was eaten is recorded separately from what was bought: half a portion of fries counts as half. Dependency-free local Python scripts build a combo report, monthly and yearly recaps as a web page and a 1080 by 1780 long image, total spending and check nutrition against that day's target. An offline demo, `python -X utf8 demo.py`, needs no token.

Who it is for. People who train on a plan, often order at McDonald's in China and want to compare combo prices and nutrition and keep a personal ledger.

Good fit when

  • It is a training day and you need a combo with enough protein that keeps your favorite burger
  • You want to swap a drink or side for a cheaper option and see the price, calorie and macro change
  • At month end you want a page with spending, favorite items and target adherence plus a long image to share
  • You need to see how the training cycle shifts if a day is skipped

Not a fit when

  • You want a medical meal plan: demo targets are not personal prescriptions and the user sets the targets
  • You want the skill to pay: the user pays through the official channel and the skill only creates the order
  • You need other fast-food chains or Russian restaurants: only McDonald's China is supported

Example request

It is a training day. Pick lunch, keep my cheeseburger, compare combos by price and macros, and do not order yet.

Limitations

A personal token for the official McDonald's China MCP is required, connected on your own machine and never committed to the repository. The demo and local scripts work without a token. The project is unofficial, built for a developer contest and unrelated to McDonald's. Real payment, coupon redemption and delivery, group and reservation orders have not been verified live, though the parameter chain is covered by tests. DeepSeek Agent has not been tested end to end. Prices hold only for the combos that were queried, and product images belong to their owners and fall outside the MIT license.

How to disable. Delete the skill folder from the agent's skills directory and the package uploaded to WorkBuddy. Disconnect the MCP connector and revoke the token on the McDonald's open platform. The local ledger lives outside the skill folder; delete it by hand.

FAQ

How do I try it without a token?

Clone the repository, then run `python -X utf8 demo.py` in the folder. You need only Python 3.11 or newer, and the demo uses synthetic menus, prices and targets and orders nothing.

How do I install it in WorkBuddy?

Download `mcd-fit-planner-workbuddy.zip` from the releases, choose add and upload in WorkBuddy's skills section, connect the connector to `https://mcp.mcd.cn` with your own token and check the link with a menu query.

How do I connect it to Codex?

Put the skill folder from the general package into `~/.agents/skills/` or the project's `.agents/skills/`. Take the MCP settings from `examples/codex-config.example.toml` and set the token in `MCD_MCP_TOKEN`.

Does the skill place orders on its own?

No. Before an order it shows price, components and nutrition, reprices after your confirmation and creates the order once. You pay through the official channel, and an unpaid order does not count as spending.

Where is the ledger stored?

In a fixed local folder outside the skill directory. Every command takes `--db` with the path to the `planner.sqlite3` file so data survives skill updates.

Discussion

No questions or reviews yet.

Skills

Skill for McDonald's China by calories and protein: computes nutrition per ingredient and finds the cheapest combos

SkillBroad access123

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 access1.3K

A skill that plans a China McDonald's order under a budget and sodium limit, checks store coupons and never orders.

SkillMinimal access582

A skill calculates daily calorie targets and builds three ready McDonald's China meal sets from the menu, with sodium warnings.

SkillBroad access104

Skill for McDonald's China orders: enumerates combos under budget and calorie caps and returns three plans

SkillExtended access124

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

SkillExtended access208

README in short

The README is in Chinese. It shows the skill remembering a plan, comparing combos and keeping an order ledger and monthly recaps. Then come the offline demo, installation in WorkBuddy, Codex and DeepSeek Agent, the combo report, monthly and yearly recaps, sample prompts, a process diagram, checks and a note on the project's unofficial status.

README badge

Original SKILL.md

---
name: mcd-fit-planner
description: 根据循环训练日与营养目标推荐麦当劳搭配，比较套餐价格和营养，记录实际摄入与消费，生成月末年末吃麦网页及分享长图。用户要按计划吃麦、顺延训练日、查账本或回顾吃麦与健身目标时使用。
---

# 麦当劳饮食计划助手

用本技能内 `scripts/fit_core.py` 完成计划、营养计算与账本操作。以技能文件所在目录解析脚本路径，所有调用输出 JSON。Python 3.11+，标准库，无本地网络客户端。MCP 通过宿主 Agent 的官方连接器调用。

首次配置和命令见 [references/local-data.md](references/local-data.md)；推荐/下单时读取 [references/mcp-flow.md](references/mcp-flow.md)。联网补充缺失营养与最终报告见 [references/nutrition-report.md](references/nutrition-report.md)。在 WorkBuddy 安装与首次只读验证见 [references/workbuddy-setup.md](references/workbuddy-setup.md)。先确认 Python 3.11+ 可用。本地库默认技能目录下 `data/`；为跨技能升级保留账本，首次在用户选择的固定目录指定 `--db <绝对路径>` 并保存该路径，之后统一使用。更换数据库路径时先确认是否是用户原有账本；不要无意建立多个真实账本。

Report a problem