AI assistant · N5 Complete system
Training Deck Builder for Restaurants
Pick the role above, how long the session runs and how often it repeats, and the assistant builds THAT deck: a slide file ready to project, with the facilitator's word-for-word script and the quiz that says who understood. Length rules: fifteen minutes hold one behaviour, ninety hold three and a drill.
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
- So training stops depending on you having a free afternoon.
- Pick a topic — how a table is greeted, how the loin is portioned, how the register closes —, the assistant turns it into a projectable 12-to-15-slide deck with a word-by-word script for whoever facilitates, and closes with a short quiz that says who got it and who needs a rerun.
- The whole session fits in 30 minutes of a slow Tuesday.
What you end up with
A working capacitacion.html file: 12 to 15 projectable slides (keyboard navigation), a word-by-word facilitator script per slide, the 5-question quiz with answers, and the session log sheet; plus the 30-minute session plan and the rerun rule for whoever doesn't pass.
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: «I need the table-greeting training for Sazón de Origen's new servers: the greeting, the timings and what is NEVER done.»
Behavior objective: greet the table in 60 seconds with the house welcome and take the order ticket without asking twice.
| # | Slide | Behavior it builds | Minute |
|---|---|---|---|
| 1 | Why it matters | Recognizing the 60 seconds in which the guest decides to return | 0-2 |
| 3 | The house greeting | Saying it whole and in the house's tone | 4-7 |
| 5 | The timings | Table greeted in 60 seconds · menus in 2 minutes | 8-11 |
| 8 | What is NEVER done | The house's 3 real mistakes, no personal names | 13-16 |
| 11 | Pair practice | 2 rounds of 3 minutes; one plays the guest | 17-23 |
| 13 | Quiz | Five behavior questions, not theory | 24-28 |
| 14 | Close | Offering the star dish up front, without reciting it | 29-30 |
SUPUESTO: to estimate what NOT training costs, the server turnover cost declared by the house is used, not an industry figure. If the real one were double, the 30-minute session pays for itself with a single person staying one month longer.
14 slides, a single file, arrow-key navigation. Under each one, the word-by-word script, visible only in facilitator mode.
| Piece inside the file | What it carries | |||
|---|---|---|---|---|
| Slides 1-12 | Title, image or diagram, and the behavior written at the foot | |||
| Facilitator script | Running text with the pauses marked: «ask here who does it differently» | |||
| Timed practice | A 3-minute clock on screen, 2 rounds, with each person's part | |||
| Five-question quiz | The right answer and its reason under each one | |||
| Log sheet | Rows of Name \ | Score \ | Verdict \ | Rerun date |
One quiz question, exactly as it lands: «A family with two kids walks in and you are carrying three plates. What do you do first?» — the right answer is to look at them and say «I'll be right with you», because the greeting costs 2 seconds and the perceived wait is cut right there.
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 30-minute plan** with an owner per block, the **rerun rule** at 7 days with its per-person verdict, and the monthly coverage review per location with whoever reads it.
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 Team size. 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.
v1.2.02026-08-27Diego's brief on the two c23 cards: «they must have a role selector and selectors for time and cadence so they land powerful». It was published with not a single axis declared: it promised choosing the topic and offered nowhere to choose anything. Role, session length and cadence come in, and the `duracion_sesion` and `frecuencia` axes are born with it.v1.1.02026-08-25Body rewritten to pay down duelo debt: it scored 54.4 out of 90.0. Two signals read ZERO, and not for lack of craft: `lexico`, because the body talked about «training» and never named the restaurant task being trained — order ticket, portion weight, portioning, waste — and `de_a_una`, because the TASK asked for everything in one block of questions. The method also ran as a paragraph («First… Second…»), so `metodo` counted 3 of 6 with six steps hidden in prose: it is now six numbered steps injecting «nombre_restaurante», «tono_voz», «meta_principal» and «tamano_equipo» inside the TASK, which is what arraigo requires and what the previous version did in no working section at all. Deliverable pieces are named by block, five BEFORE YOU DELIVER checks are written out, and a SUPUESTO line is added so the AI declares its own turnover figure instead of the house hard-coding one. Rises to 97.5.v1.0.02026-08-14Initial version. Born in the 🛠 tools line (artifact-titles batch 1), first published card in c23, with the full lead-magnet kit and a sample from the fictional restaurant Sazón de Origen.More assistants in this category
Pick the role above and which door the person comes through, and the assistant writes the whole process from a…
N5 · Complete systemYou pick the stage and the role, and the assistant designs training backwards: the test you can fail first, th…
Included in the always-growing library
Access to every published assistant, adapted to your AI and personalised with your restaurant's data, updates included.
