AI assistant · N5 Complete system
Team Knowledge to Written Procedures Converter for Restaurants
It gets out of the head of whoever has been doing a task well for years the step-by-step nobody ever wrote down, and leaves it as a written procedure with its standard, its check and its owner —before that person leaves—. Choose above where it comes from: an interview with the team, a recording or a voice note.
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 levelIs this for you?
What it does and what you get
- The most expensive thing that walks out when someone quits is in no manual: how Marcela negotiates with the difficult supplier, how she squares the register when it won't square, how Andrés cools down a heated complaint.
- That knowledge can't be filmed or voice-noted — it lives in decisions and criteria, not movements.
- This assistant extracts it with the apprentice interview: asking like someone doing the task tomorrow — what do you do first?
- what can go wrong?
- how do you know it came out right? — and turns the answers into an SOP with steps, decisions, common mistakes and done-right criteria. The starting rule crosses with the people board: you begin with the knowledge where the house depends on ONE person.
What you end up with
The critical-knowledge map prioritized by single-person dependency, with verdict; the apprentice interview tailored to the chosen knowledge; and the SOP built from the answers — steps and decisions, common mistakes with their fix, done-right criteria — with the expert's validation, a version and the monthly capture routine.
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
The owner asked this: «Marcela is the only one who can square the register when it won't square, and she's been here 6 years. If she leaves, I'm blind. I want that captured now.»
| Knowledge | Who holds it | What breaks without it | Backup | Starting verdict |
|---|---|---|---|---|
| Register squaring with a gap | Only Marcela | Every unresolved gap ends as a loss or an unfair suspicion | No | FIRST — single person, weekly money |
| Negotiating with the pork supplier | Only the owner | The main ingredient's food cost climbs | No | Second: captured next month |
| Heated complaint on the floor | Andrés, partly Marcela | Reviews and lost tables | Partial | Third — backup is only half there |
SUPUESTO: no industry figure is hard-coded for how long training a register backup takes. Priority is ordered by what Sazón de Origen moves in cash each week; if the owner brings their own data, the order is recalculated in front of them.
Record it or take notes, without interrupting. Under each question sits its why follow-up.
- Tomorrow I square the register: what do I do first, before counting a single peso? (why in that order?)
- How do you separate counted from expected, and what do you check first when it won't square? (why start there?)
- What are the 3 most frequent causes of a gap here? (why do they repeat?)
- What do you do differently when there is extra money instead of missing money? (why does it worry you just as much?)
- How do you treat voided tickets and delivery orders charged outside the register? (why are they counted apart?)
- When do you stop searching and log it as a loss, and who do you tell? (why that ceiling?)
- What mistake did someone once make that you no longer make? (why was it being made?)
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 are **the SOP with its decision tree** and **its governance**: the cover signed by the author, the done-right criteria with units, the open questions with owners, and the monthly review checklist with its numeric target.
See the full example — free accountWorks with these AIs
How to use it
Which of your data it uses
This assistant works with 8 of the 23 fields in «My Restaurant», among them Restaurant name, Business type and Cuisine or concept. 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.
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.
v1.2.02026-08-27It was called «Convertidor del Conocimiento del Equipo en SOP para Restaurantes». Diego's instruction, 2026-08-27: «this name is not understood, what is a lamina? if it is not understood it does not get used». A title that has to be translated does not get opened, and what does not get opened does not exist. The assistant's body does not change: what changes is the word it is searched and found by.v1.1.02026-08-25Body rewritten to pay down duelo debt: it scored 62.8 out of 90.0. Two signals read ZERO: `cifras`, because a card that orders an interview and sets a target carried no measurable window at all —the «twenty minutes» and «six months» were spelled out in words—; and `lexico`, because the assistant talked about knowledge in the abstract without naming what knowledge a restaurant lives on: the register squaring with its voided tickets and delivery orders, the negotiation that holds the main ingredient's food cost, the standardized recipe an SOP does NOT replace, and the portion weight or waste ceiling that turn «make it good» into an observable done-right criterion. `verificacion` sat at 1 for the same reason as the rest of the corpus: BEFORE YOU DELIVER was a row of questions with no checking verb. And the TASK ran as prose, so `metodo` counted 3 of 6; it is now six numbered steps, five of them injecting the restaurant's «fields», which is what arraigo requires since its cutoff —the eight fields were declared under YOUR RESTAURANT and never appeared again—. A SUPUESTO line is added so the AI declares its own benchmark. Rises to 95.0.v1.0.02026-08-15Initial version. Born in batch 11, third of the c26 family: the video one films the task, the audio one records the voice, and this one INTERVIEWS the knowledge that can't be filmed — decisions and judgment. With the full lead-magnet kit, a Sazón de Origen example and a written cross-link to the people board's cross-training matrix.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.
More assistants in this category
Pick the area above, WHICH manual you want and the size of your operation: the processes, the house culture, i…
N5 · Complete systemPick the form above —opening, closing, goods-in, temperatures, cleaning, till, inventory or maintenance— and t…
N5 · Complete systemRecord a voice note explaining how the task is done —while doing it— and walk out with the step-by-step manual…
Included in the always-growing library
Access to every published assistant, adapted to your AI and personalised with your restaurant's data, updates included.
