Who is this assistant for?

The dish only comes out right when the usual cook is in, and you know it because complaints land on the Tuesdays he is off. The recipe you have says "a cup of rice" and "salt to taste", and nothing can be weighed with that. The hogao is copied inside five different recipes, so a supplier change means correcting all five. Nobody knows how many portions come out of the big pot or how many days the leftovers hold.

When a dish only comes out right if the usual cook is in, the house does not have a recipe: it has a dependency. This assistant turns the preparation into a document a new cook can run without asking, with the batch declared, the net weight of every ingredient and a control point on every step. It does not cost and it does not plate: it fixes how the thing is made, which is what later makes costing and plating possible. And once the recipe is standardized and the busy day asks for ninety portions, it scales it to the production batch with kettle judgment: the factor only on what scales linearly, salt and liquid closed in a pilot batch, times by thermal mass and the runs that fit the equipment.

AI assistant · N5 Complete system

◆ Included in the annual plan👨‍🍳 Standard Recipes, Spec Sheets and Productionv1.2.0 · 2026-09-24

📚 Works best with an AI that reads long documents in one pass (your full menu, a manual) · optional

Standardized Recipe Builder for Restaurants

A standard recipe and a recipe card are the SAME document with two names, and at the top you pick what your kitchen calls it. In this one —standard recipe— production leads: writes each preparation's standardized recipe as a production document: batch, declared yield, net weights, method with control points and storage, proven against one batch actually weighed.

Depth

9/23

Your data

Compatible AIs

Steps to use it

Depth of the resulttap a level to see what you get

◀ Put out today's fireRun the year ▶

🏛️

N5Complete system

Several chained pieces you install and operate over time, with their review cadence. It changes how the business runs, not one decision.

◆ This assistant 110 of the 164 published assistants are at this level

Is this for you?

What it does and what you get

When a dish only comes out right if the usual cook is in, the house does not have a recipe: it has a dependency.
This assistant turns the preparation into a document a new cook can run without asking, with the batch declared, the net weight of every ingredient and a control point on every step.
It does not cost and it does not plate: it fixes how the thing is made, which is what later makes costing and plating possible.
And once the recipe is standardized and the busy day asks for ninety portions, it scales it to the production batch with kettle judgment: the factor only on what scales linearly, salt and liquid closed in a pilot batch, times by thermal mass and the runs that fit the equipment.

What you end up with

The recipe header with batch, portions, portion weight and shelf life; the ingredient table in net weight with its measuring tool, the cited sub-recipes and the method with one control point per step; plus the weighed-batch proof against the declared batch, the new cook's checklist and the recipe-book coverage indicator. If you choose to scale it: the batch sheet with its factor, the scaling table by family with the reason for each adjustment, the capacity matrix with the runs, the production sequence and the pilot batch sheet with the measured yield.

EXAMPLE This is what you getAn example built with Sazón de Origen, the house's fictional restaurant.Not your data: your result is built with YOUR restaurant's.See the full example →

Sample result

Sazón de OrigenFictional contemporary Colombian restaurant in Medellín, 2 locations, 6 years in business. Every sample result in the library comes from this same case. Fictional restaurant

The owner asked this: "Marta makes the house hogao and she makes it from memory. When Marta is off, the hogao comes out different and you can taste it in the bandeja and in the sancocho. Write it down so anyone can make it."

Preparation	Type	Batch size	Portions produced	Portion weight	Shelf life	Version and date	Owner
House hogao	Mise en place sub-recipe	1 pot · 7.05 kg in	70	45 g	5 days at 4 °C	v1 · 26-Aug	Marta Q. (production chef)

The header sits on top and in bold because it is the one thing nobody changes without telling anyone. If one day 62 portions come out instead of 70, the paper is not patched by hand: you weigh again and you raise the version.

And one house decision worth stating: this recipe enters the bandeja, the sancocho and the trout BY CITATION. None of those three dishes carries the hogao written inside it.

Ingredient	Net weight	Unit	Measuring tool	Prep note
Ripe chonto tomato	4,200	g	floor scale	cored, in 1 cm dice
Spring onion	1,500	g	floor scale	white part plus 2 cm of green
Sunflower oil	350	ml	graduated jug	—
Peeled garlic	60	g	gram scale	crushed, not chopped
Annatto colouring	12	g	level 6 g spoon × 2	dissolved in warm oil
Salt	55	g	gram scale	half at the start, half at the close
Ground cumin	4	g	level 2 g spoon × 2	—

Cited sub-recipes: none. This IS the sub-recipe.

Method

Heat the oil with the annatto over medium heat, 3 min, until the oil takes an even orange tone. Control point: if it smokes, it goes in the bin and you start again.
Sweat the onion 8 min, stirring every 2 min. Control point: translucent, no colour at the edges.
Add the garlic, 1 min. Control point: perfume, not colour.
Add the tomato and half the salt. Cook 35 min on low, covered for the first 15 min. Control point: the tomato loses the shape of the dice and releases its water.
Uncover and reduce 12 min. Control point: the wooden spoon leaves a track on the bottom and the track takes 3 seconds to close.
Finish with the remaining salt and the cumin. Control point: taste; if it needs more salt, how much was added goes on the paper, it is not added in silence.

Storage: in a covered 4 litre tub, labelled with preparation, date, time and owner. 5 days at 4 °C. Reheat in a water bath, never straight in the microwave, because the oil splits and the hogao looks broken on the bandeja.

Minimum equipment: 12 litre pot, gram scale, 4 litre tub. Site 2 has no 12 litre pot: there the recipe runs at half batch and yields 35 portions, and that is written into its own version of the header.

Illustrative example generated with a fictional restaurant. Not a promise of results, not a client case.

This is 2 of 3 parts. Behind the wall is **the part that turns paper into a standard**: the weighed-batch proof against the declared batch with waste split into trim and cooking loss, the new cook's 6-point checklist, what changes in portioning when the dish goes out for delivery, the recipe-book coverage indicator with its fixed day, and the blank template for the next sub-recipe.

See the full example — free account

Works with these AIs

Claude — the best for this assistant The best one here because it does not fill gaps: when you do not know an ingredient's weight it asks instead of inventing, and it separates trim waste from cooking loss on its own. Ask it to leave the gap flagged as pending rather than guess a number.

How to use it

Pick at the top what your kitchen calls it. Same tool: as a standard recipe, production leads —batch, yield, procedure, times—; as a recipe card, the pass and the floor lead —plating, photo, weight per component, allergens, price—. If your house keeps them as two papers, both come out of one source.

Also choose what you do with the recipe: standardize it (fix how today's batch is made) or scale it to a production batch. Scaling needs a standard recipe: if you don't have one, the assistant standardizes it first.

Fill in My Restaurant and pick the preparation to start with: the best seller, or the one only one person knows.

Paste the assistant into your preferred AI and answer its questions on batch size, portions and the actual tools at the station.

Print the draft and produce one batch following it to the letter, scale beside you.

Correct the paper with whatever the scale said, then sign it with date, version and owner.

Repeat with one preparation a day for 2 weeks and weigh every batch again after 30 days.

Which of your data it uses

of 23

This assistant works with 9 of the 23 fields in «My Restaurant», among them Restaurant name, Cuisine or concept and Number of locations. It does not ask out of curiosity: these are what make the answer speak about YOUR scale instead of an industry average — the same calculation on a business of another size returns a number you cannot decide with. You fill them once and they apply across the library, so the second assistant you open already starts with them in place. Whatever you leave blank, the assistant asks for one thing at a time instead of inventing it, and says so before answering, so you know what it is working from.

Restaurant nameCuisine or conceptNumber of locationsTeam sizeBest-selling dishesStrong and slow daypartsActive sales channelsApproximate food cost (%)Main goal this year

Always up to date

Every assistant carries a visible version and date. When an AI changes how it works, the assistant's version goes up and you see what changed.

Same method, another variant

Inside the app this is a single control: you press it and the assistant reframes itself, including the variants that don't appear here because they don't have their own page yet.

What your kitchen calls it

standard reciperecipe card

What your kitchen calls it

standard reciperecipe card