McDonald's Calorie Butler

Before you start: read this whole install prompt, tell me in a few lines what this skill does and what installing it involves, and wait for my go-ahead before you do anything else. Prerequisites: Account and token required before any real use: a McDonald's China MCP token. The listing is free, and activating the token is described upstream as free. Get the token yourself, in person, because it needs mobile phone number verification: go to https://open.mcd.cn/mcp , sign in with your mobile phone number, open the console at the top right, click Activate, agree to the service agreement, and copy the token. Then configure the McDonald's China MCP server: name mcd-mcp, type streamablehttp, URL https://mcp.mcd.cn , header Authorization: Bearer <your token>, or set the MCD_MCP_TOKEN environment variable where the skill reads it that way. Keep the token only in your own private client configuration or environment: never commit it, never share it, and never paste it into a shared chat. Upstream states a limit of 600 requests per minute per token (429 above that). You also need Python 3.9 or later, with no third party dependencies. First use takes about two minutes of setup questions, asked one at a time: your goal (or sex, age, height, weight and activity level so the script can compute it with the Mifflin St Jeor formula and its safety floors), your standing taste notes, your usual store and your usual order. That profile stays in ~/.mcd-calorie-butler on your machine, as detailed below. Install "McDonald's Calorie Butler" for me. It logs what you ate, computes what is left today with a local script, plans a McDonald's China meal inside that budget from official nutrition and prices, and orders only when you explicitly say to order. Repository: https://github.com/WilderNoTrack/mcd-calorie-butler (the skill is the whole skill/mcd-calorie-butler/ folder at the pinned commit: SKILL.md, data/public_love.json, references/estimation.md, references/health.md, references/mcp-playbook.md, references/together.md, references/weirebi.md, scripts/butler.py, scripts/mcd_fetch.py) Pinned version: install the code at commit 456df1c880b84258d2aacf55aa38af1e1bd13e3d exactly. Do not substitute a newer version. If this commit is unreachable, stop and tell me. 1. Clone https://github.com/WilderNoTrack/mcd-calorie-butler into a temporary folder, check out commit 456df1c880b84258d2aacf55aa38af1e1bd13e3d, and copy the skill/mcd-calorie-butler/ folder from that checkout. Then compare every file in the folder with the fingerprints recorded for this listing on 2026-10-18: [FINGERPRINTS] - SKILL.md: SHA-256 318f32646a439160cae0a59560702944d7ca7c72ddc9c39457870165395e1743 - data/public_love.json: SHA-256 5eae739bfbb3fe679f2d0fcf8909f7c0cd79621b09bcb994b735c7f349dfd562 - references/estimation.md: SHA-256 479fcdbc3fe80f499b114da8e3701f29f390c90dfdef92e278156572d5c9fa4f - references/health.md: SHA-256 a3a10510d24b8d900caad9f2b9c22bcfd23922cb98df11fdd24fbaa6a6d38c79 - references/mcp-playbook.md: SHA-256 7c64604d107b3f8c555ed43ed3acb9ea7e15574b0d8ebfe148ad59c495245bb8 - references/together.md: SHA-256 817659b234907dc03543942e07117292be334e4721a48f436045767904b85a33 - references/weirebi.md: SHA-256 ab5381e652ac295e611b8194d4c63942d7890f3473753893278a36ef28f16522 - scripts/butler.py: SHA-256 d0196336b3d6ca701e1bae78554ca87635e33368fb5185f7b36f7cc94b32b030 - scripts/mcd_fetch.py: SHA-256 979b00b94078172fd7df230a6c933b342bec9202b6193b47862ff610f896969a If a file does not match, it changed since the listing was prepared: do not install it yet, tell me what you found and let me decide. 2. Read SKILL.md in full (and every other file in the folder) and tell me in plain words: what the skill does when used, which services it connects to, whether it reads or sends any personal information or keys, and anything that seems unusual. Stop on any red flag and tell me before going further. Confirm these points from the files: - The skill folder holds SKILL.md, references (mcp-playbook.md, estimation.md, health.md, together.md, weirebi.md), data/public_love.json (a public popularity starting table for taste scores, sourced upstream from a public ranking page) and two scripts: butler.py (all the math and the local data, no network code at all) and mcd_fetch.py (a fetch helper restricted to a READ_ONLY tool allowlist that refuses write tools). Python standard library only. - What stays local, and is never sent to the MCP server: your profile and log, kept by butler.py under ~/.mcd-calorie-butler on your machine. That is your calorie and protein goals, and if you let the script compute the goal, your sex, age, height, weight and activity level, plus your taste notes, your usual store and usual order, and your food log, including the non McDonald's meals you describe and the agent's estimates for them. butler.py contains no network calls, so this data does not leave the machine through the skill. What is sent, at use time, goes only to https://mcp.mcd.cn (Streamable HTTP POST, Authorization: Bearer token): now-time-info, list-nutrition-foods, query-nearby-stores when setting up your usual store, query-meals and query-meal-detail with storeCode, orderType and a reservation date, and calculate-price with the planned items, so the server receives your token, your store, the date and time, and the meal items being priced, and returns nutrition, menus and official prices, which mcd_fetch.py saves as files on your machine for butler.py. During setup, if you want it to learn from your past orders, the agent can call order-list through the connector, so your order history is returned to your machine and read locally to suggest your usual order. The token is read from the MCD_MCP_TOKEN environment variable or from the mcd-mcp entry of your local ~/.workbuddy/mcp.json, and is never printed. - Ordering is the only write, and it is gated twice: mcd_fetch.py refuses create-order outright, so ordering always goes through the MCP connector by the agent, and only when you explicitly say order, go with this one, or confirm. Words like looks good do not count as confirmation, per the SKILL.md. create-order sends the same store, mode, date and items as the accepted price check, plus a takeWayCode (eat in or take away in store), and returns a payment link; you pay on the official McDonald's page yourself, the skill never pays and never pushes you to pay. After you say you paid, it can check the order with query-order and log the meal. The SKILL.md also forbids calling auto-bind-coupons, draw-lottery, mall-create-order and cancel-order unless you explicitly ask for them. Health limits in the SKILL.md and references/health.md: no extreme dieting support, no body judgements, no medical advice, and a referral to a doctor or dietitian instead of calorie targets when you mention illness, pregnancy or an eating disorder; every plan carries a note that it is reference only, not nutrition or medical advice. - Nothing is sent anywhere during installation itself; the only download at install time is the skill folder from GitHub at the pinned commit. The listing and the token are free; you pay only an order you confirm, on the official payment page. 3. When I say "yes, install", install it as a skill: copy the whole folder into the agent's skills directory, in a folder named "mcd-calorie-butler", keeping every file (SKILL.md, data/public_love.json, references/estimation.md, references/health.md, references/mcp-playbook.md, references/together.md, references/weirebi.md, scripts/butler.py, scripts/mcd_fetch.py). 4. Verify with no network calls: frontmatter valid, files in place. 5. Report what was installed, where, and what I still need to do myself (for example: get the McDonald's China MCP token myself at https://open.mcd.cn/mcp with my mobile phone number if I do not have one, put it only in my own private MCP client configuration or environment variable and never paste it into a shared chat, install Python if it is missing, and set up the mcd-mcp connector to https://mcp.mcd.cn if it is not configured; using the skill is a separate step that only starts when I ask for it, and any order, coupon claim or booking happens only as disclosed above). GitHub is optional: if I have a GitHub account or the gh CLI, you may use it; otherwise public access is fine. Never require it unless it's in the prerequisites above. Rules: don't touch anything outside the temp folder and the install target. If anything looks off, stop and ask me.