McDonald's Saver Strategist

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 3.8 or later to run the optimizer and the other scripts (standard library only; Node.js is only needed for the optional scripts/parity_check.mjs). The skill can also work from sample data offline (scripts/sample_data) with no token, but any real menu, coupon or order data needs the token and connector above. For delivery you need a delivery address on your McDonald's account, or to add one when asked. Install "McDonald's Saver Strategist" for me. It pulls your store menu and every coupon pool from the McDonald's China MCP, runs a deterministic local optimizer to find the cheapest valid order, and orders only after your explicit confirmation. Repository: https://github.com/hujjun/mcd-saver-strategist (the skill is the whole skills/mcd-saver-strategist/ folder at the pinned commit: SKILL.md, references/coupon-strategy.md, references/mcp-tools.md, scripts/bench.py, scripts/demo.py, scripts/mcp_probe.py, scripts/optimizer.py, scripts/parity_check.mjs, scripts/sample_data/coupons.json, scripts/sample_data/menu.json, scripts/sample_data/request.json, scripts/selftest.py) Pinned version: install the code at commit 1f8ebbea06c24fd4731405ed86894c577b40fc71 exactly. Do not substitute a newer version. If this commit is unreachable, stop and tell me. 1. Clone https://github.com/hujjun/mcd-saver-strategist into a temporary folder, check out commit 1f8ebbea06c24fd4731405ed86894c577b40fc71, and copy the skills/mcd-saver-strategist/ 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 caed9c6cbb79855af700a45b611918022eaef3928239ef45d52880a39e8803b2 - references/coupon-strategy.md: SHA-256 a42e8ba06f7c2cc78ca34c38eed087c46dadbd270ad65111fc80daa23bc301d2 - references/mcp-tools.md: SHA-256 025717ad137bc59fa228a4b3b7093500dd5573d25d053de989a48cb06b8155c2 - scripts/bench.py: SHA-256 7a150e594a2da74545bdee0bc4db90ba1836a544518b309151b1799b42afa912 - scripts/demo.py: SHA-256 b0b21c0ce87fb190678e17cfa1aca0afb2b0d767a3916cf790d82779748a1159 - scripts/mcp_probe.py: SHA-256 0f3abc35ccce684cb679d202f0110b304ecb30773995dccf2ff8ebee914647a7 - scripts/optimizer.py: SHA-256 ab120292475154038f747566b87b9274be9e1aed49363d8931924fecb4338b4d - scripts/parity_check.mjs: SHA-256 700a12f64097ba55c042bd5853f0f33e73c68c212f0a5dafd62a03ae96ae9898 - scripts/sample_data/coupons.json: SHA-256 74c54fac90e0c307c9aff0563b6622aafbf90579a745abe9635ab47c224f9ef3 - scripts/sample_data/menu.json: SHA-256 4e69f8148086c6c3216ed495179b12cd37c7cd20a8120933da98cd38261d7a52 - scripts/sample_data/request.json: SHA-256 99214c2409a602aa529a2d4989b10e8833bba877d2666e0656e4adfae8bdc347 - scripts/selftest.py: SHA-256 64c76b4ffa403b289c05a9f6087283c1932496358e334119c126e1205a52d066 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, two reference files (references/mcp-tools.md and references/coupon-strategy.md) and scripts: optimizer.py (the deterministic optimizer, Python standard library only), selftest.py, demo.py, bench.py, mcp_probe.py (an MCP connectivity check), parity_check.mjs (optional, needs Node.js) and sample_data (coupons.json, menu.json, request.json) for offline runs. No other code and no dependency list of its own. - At use time, nothing is sent anywhere except to the McDonald's China MCP server at https://mcp.mcd.cn (Streamable HTTP, JSON-RPC POST, Authorization: Bearer token on every call). The calls send, in this order: a time query (now-time-info); store lookup for pickup (query-nearby-stores, with the city or keyword you give) or, for delivery, a read of the delivery addresses already saved on your account (delivery-query-addresses) then delivery-query-stores; query-meals and query-meal-detail for the chosen store (storeCode and item codes); the four coupon queries (available-coupons, auto-bind-coupons, query-my-coupons, query-store-coupons); list-nutrition-foods only when you set a calorie or nutrition limit; and calculate-price with the chosen item codes for the final price check. The server therefore receives your token, your store choice, any city or keyword you give it, your item selections, your budget and calorie limits as tool arguments, and it returns your account coupons and saved addresses to your machine. The local optimizer only reads local JSON snapshots (menu, coupons, request) written to a temp folder on your machine; it makes no network calls itself. The only script that calls the server directly is mcp_probe.py, which sends the same Bearer token and a static User-Agent naming this project and its GitHub address. - Two write actions on your McDonald's account are part of this skill, and both must be disclosed before they happen. First, auto-bind-coupons: the SKILL.md workflow treats it as a required step, so during a normal savings run the skill claims every claimable McMaiSheng coupon to your account, without a separate consent step in the upstream workflow. Second, ordering: create-order is called only after the agent repeats the final price, store and item list to you and you explicitly confirm; it returns a payment link that you pay yourself, and cancel-order is used only if you ask to cancel. The upstream tool list also includes delivery-create-address, used when you have no delivery address: it sends the new address you provide, with its city, contact name, gender and phone number, to McDonald's. - Nothing is sent anywhere during installation itself; the only download at install time is the skill folder from GitHub at the pinned commit. Using the skill is not charged by the skill itself: you pay only the McDonald's order you confirm, at the official price returned by calculate-price. 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-saver-strategist", keeping every file (SKILL.md, references/coupon-strategy.md, references/mcp-tools.md, scripts/bench.py, scripts/demo.py, scripts/mcp_probe.py, scripts/optimizer.py, scripts/parity_check.mjs, scripts/sample_data/coupons.json, scripts/sample_data/menu.json, scripts/sample_data/request.json, scripts/selftest.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.