About dishmath
dishmath is a local restaurant-calculation server for restaurant operators, covering ingredient costs, VAT-aware menu pricing, recipe scaling, and EU allergen checks. It computes results on the machine and exposes four calculations through MCP and a CLI.
Key features
- 01Ingredient-level food cost calculations
- 02Menu pricing with VAT and rounding
- 03Recipe quantity and cost scaling
- 04EU allergen matrix for 14 allergens
Use cases
- 01Calculate a dish’s cost and target selling price
- 02Set a VAT-inclusive menu price
- 03Scale ingredient quantities for new portions
- 04Identify dishes to avoid for an allergen
MCP support
Exposes `food_cost_calculator`, `menu_price_calculator`, `recipe_scaler`, and `allergen_matrix` to MCP clients. Calculations run locally from supplied inputs.
Configuration
Local server config is the same for every client. Copy it into your client’s MCP config file.
Tools
4Saved tool list · updated 1w ago
| Parameter | Type | Required | Description |
|---|---|---|---|
| currency | string | No | Currency code, e.g. USD. |
| ingredients | array | Yes | The ingredients for this call. |
| current_price | number | No | Current price value. |
| target_food_cost_percent | number | No | The target food cost percent for this call. |
Details
From the project READMEdishmath
Restaurant arithmetic that runs entirely on your own machine: what a plate costs to make, what to charge for it, how a recipe changes between portion counts, and which dishes a guest with an allergy must avoid. There is no account, no key and no network call — every answer is computed locally from the numbers you pass in.
The four calculations
- Food cost — add up a dish ingredient by ingredient. Alongside the plate cost you can ask for the selling price that lands on a target food-cost percentage, and for the percentage you are really running at today's price.
- Menu price — start from a plate cost and a target food-cost percentage. You get the net price, the price with VAT added, that price rounded up to a clean step (0.50, 1.00, …), the gross profit per plate net of tax, and the food-cost percentage the rounded price actually gives you.
- Recipe scaling — give a recipe written for one number of portions and get every quantity rewritten for another. When the ingredients carry unit costs, the batch cost and the cost per portion come with it.
- Allergen matrix — declare the allergens in each dish and get the dish-by-allergen table used under EU Regulation 1169/2011 (Annex II), plus the list of dishes to avoid for each of the 14 allergens. Names are accepted in English or Italian, and common names work too: "shrimp" counts as crustaceans, "dairy" as milk. The tool never guesses allergens from a dish's name — a dish with none declared is reported that way.
Money is rounded half-up to two decimals and percentages to one
decimal. An optional ISO currency code (default EUR) changes only
how amounts are printed.
Worked example, plate cost €3.20 at a 30% target with 10% VAT and a 0.50 rounding step:
Net price (before tax): €10.67
Price incl. 10% tax: €11.74
Menu price (rounded up to 0.5): €12.00
Gross profit per plate (net of tax): €7.71
Real food cost after rounding: 29.3%
CLI
npx @reneturcios/dishmath menu_price_calculator '{"plate_cost":3.20,"target_food_cost_percent":30,"vat_percent":10,"rounding_step":0.50}'
npx @reneturcios/dishmath food_cost_calculator '{"ingredients":[{"name":"flour","quantity":0.5,"unit":"kg","unit_cost":1.20}],"target_food_cost_percent":30}'
npx @reneturcios/dishmath recipe_scaler '{"ingredients":[{"name":"flour","quantity":500,"unit":"g"}],"original_portions":4,"target_portions":6}'
npx @reneturcios/dishmath allergen_matrix '{"dishes":[{"name":"Shrimp pasta","allergens":["shrimp","gluten"]}]}'
npx @reneturcios/dishmath --version
Each tool subcommand takes a single JSON argument string and prints the result as plain text. Exit codes: 0 on success, 1 on a calculation or validation error, 2 on a usage error.
MCP
npx -y @reneturcios/dishmath mcp
MCP client configuration:
{
"mcpServers": {
"dishmath": {
"command": "npx",
"args": ["-y", "@reneturcios/dishmath", "mcp"]
}
}
}
Tools exposed over MCP: food_cost_calculator,
menu_price_calculator, recipe_scaler, allergen_matrix.
Docker
docker run -i ghcr.io/cloudcorprecords/dishmath
docker run --rm ghcr.io/cloudcorprecords/dishmath menu_price_calculator '{"plate_cost":3.20,"target_food_cost_percent":30,"vat_percent":10,"rounding_step":0.50}'
The container's default command is mcp, so the first form starts
the stdio MCP server; any other arguments reach the CLI.
Development
npm install
npm run build
npm test
npm run dev
License
MIT — see LICENSE.
Comments
New to MCP?
From what the protocol is to installing your first server, these answers cover most first questions.
An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.