AI assistant · N5 Complete system
Operational Checklist Builder for Restaurants
Pick the area above, the moment in the shift and how often it runs, and the assistant writes THAT checklist: the list that gets ticked with a pen at close, with its owner per line, the standard that defines «done», and what happens when something is not met. One builder for twelve areas and three periods instead of one card per combination.
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 manager ends the day exhausted and unable to say what he decided.
- This assistant separates his job from what the station already signs: it does not check fridges or portion weights, it checks the people who check, and it forces every NO on somebody else's sheet to carry an action closed inside 24 hours.
- The day's numbers always arrive with their comparison, because sales without the same weekday last week beside them send nobody to do anything.
- And the day does not close with a summary: it closes with three decisions that have an owner and a date.
What you end up with
Five chained blocks: the day's header with its target, the cross-check table of other people's checklists, the numbers table against the same weekday last week, the people table with today's conversations, and the three signed decisions with their boundary matrix.
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 have two locations and 24 people. My head chef and my floor manager sign their checklists every day, but I end up putting out fires and cannot say what I decided. I need the manager's checklist for Tuesday lunch."
Tuesday 26 August · Laureles location · Manager on duty: you · Day target: 3,900,000 COP
Tuesday-to-Friday lunch is the slot that holds this house up, so the checklist clock hangs from the 12:00 to 14:30 peak and not from opening time. The four phases fit into 70 minutes of your working day; the rest of the day you direct, you do not fill in paper.
| Phase | Hour | What you do | What does NOT belong here |
|---|---|---|---|
| Before service | 11:15 (45 minutes ahead) | Read Monday's signed checklists, count the shift's hands against the 9 people Tuesday asks for and confirm the day's target | Temperatures, portion weight and portioning: the head chef signs those |
| During the peak | 12:00 to 14:30 | Watch from the door, log turned-away tables with the hour and the exact hour each dish sold out | Do not step onto the grill or the pass, and do not answer the delivery phone |
| After the peak | 14:30 to 15:00 (30 minutes) | Take partial sales, covers served, average check and voided dockets while memory is fresh | Do not wait for close to write it: by 21:30 nobody remembers the hour |
| Close | 21:30 (15 minutes) | Sign the day's three decisions with owner, date and closing figure | Do not leave a single row on "to be reviewed" |
The second location runs the same form with the clock moved 30 minutes, because its peak starts at 12:30. Operating figures built from the data you gave, illustrative until you check them against your POS.
Three other people's checklists come into this board. None of them is the manager's: somebody else signs all three, and your job is to read what was done with each NO, not to repeat the inspection.
| Checklist | Who signs | Arrived | NO finding | Action, owner and hour |
|---|---|---|---|---|
| Kitchen opening | Head chef | Yes, 10:40 | Garlic trout prepped at 22 portions and last Tuesday 26 were sold | Raise to 30 portions from next Tuesday · Head chef · logged 11:20 |
| Floor opening | Floor manager | Yes, 11:05 | Two terrace tables unset for lack of clean linen | Linen to the laundry today 16:00 · Floor manager · closed 16:20 |
| Monday till count | Shift cashier | Did not arrive | — | Asked at 11:30, arrived 15:40 with a 12,000 COP shortfall · Cashier · signature pending |
Two of three closed inside 24 hours. The till count that came in late is the one that matters: it is not a late piece of paper, it is a till left unreconciled through four hours of service, and the shortfall only surfaced once the shift had already gone home. House rule applied here: a signature with no facts is worse than a blank box, because it reassures. The kitchen opening checklist arrived signed at 100 % and the star dish still sold out at 13:10, so the signature held and the production count was wrong: that is exactly the finding a manager sees and a station does not. Neither of those two lines is a kitchen failure; both are a planning failure that only the manager's desk can fix.
Every row is read against the same Tuesday last week. A number on its own sends nobody to do anything.
| Indicator | Today | Last Tuesday | Deviation | Reading |
|---|---|---|---|---|
| Day sales | 3,650,400 COP | 3,912,300 COP | −261,900 COP (−6.7 %) | Under the 3,900,000 COP target |
| Covers served | 78 | 81 | −3 | Three chairs, not a demand drop |
| Average check | 46,800 COP | 48,300 COP | −1,500 COP (−3.1 %) | Below the house's 48,000 COP |
| Sold-out dishes | 2 (13:10 and 13:40) | 0 | +2 | Garlic trout and three-meat sancocho |
| Turned-away tables | 4 (12:40 to 13:20) | 1 | +3 | About 9 covers unserved: 421,200 COP |
| Cancelled orders | 3 on the delivery platform | 1 | +2 | Waits over 45 minutes: 156,000 COP |
Visible leak for the day: 577,200 COP, more than the 261,900 COP missing against last Tuesday. Sales did not fall for lack of guests: they fell because the house could not serve the guests who came, and that is a capacity problem, not a demand problem. The check dropped because the two sold-out dishes were the highest priced on the menu and whoever arrived late ate the 34,000 COP option. Menu note: garlic trout sells at 42,000 COP with a 14,700 COP cost, a 35 % food cost, above the 34 % menu average and above the 32 % ceiling for promoting it; there the problem is not the offer, it is the menu price. Illustrative figures from the data you supplied, to be checked against your own POS.
Illustrative example generated with a fictional restaurant. Not a promise of results, not a client case.
This is 3 of 5 parts. Behind the wall sit the two parts only the manager can sign. The people block: one row for everyone who moved the day, with the columns Person, News, Who covers and Pending conversation, the exact hour of each conversation booked for today and the extra labour cost of the cover counted in pesos and added to the month's labour cost. And the three-decision block, with Decision, Owner, Date and the figure that closes it at 14 days, the boundary matrix listing the three things this checklist does not review and the role that does sign them, and the minute-by-minute script for the 15-minute meeting at 11:15, with the speaking order and the rule for a day with no decisions.
See the full example — free accountWorks with these AIs
How to use it
Which of your data it uses
This assistant works with 11 of the 23 fields in «My Restaurant», among them Restaurant name, Business type 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.
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.
v1.3.02026-08-27The title drops «Daily». Diego's brief: «these names must not carry monthly, because the time selector is what sets the period». And its own gate now catches it: `motor/titulo_vs_control.py` fails when a card is named after a value its chip changes IN PLACE — the owner picked «weekly» and walked away with a deliverable titled «daily». With the period axis inside, the name can only state what does not change: that this is the house's operational checklist. The URL is untouched: it stays alive and the registry keeps the old name as an alias.v1.0.12026-08-26The body stops contradicting the answer ladder. It used to say «do not move on until it arrives», the very order the ladder was born to replace: the assistant asked the owner for what the owner does not have and stalled. It now says what the house doctrine says — ask one at a time, and if they do not know, continue with a marked assumption. FORMAT OF THE RESULT was not touched, so the example still corresponds and its seal moves up with the card.v1.0.02026-08-25The assistant is born.v1.3.12026-09-21The deliverable stops being an analysis in blocks and becomes a printable OPERATING DOCUMENT: one checklist_diario.html with the print button inside, sixteen sections and THE DAY SHEET, blank and landscape. The three layers of the family arrive: the day written as a PROCEDURE with measurable standards and out-of-standard cases, the ECONOMICS OF THE DAY with the arithmetic beside every figure, and the five governance pieces.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
Builds the checklist your restaurant opens with: the walk through the venue in blocks, every box with its meas…
N4 · Executable planWrites your restaurant's closing checklist in five named phases —product, money, premises, station and shift h…
N4 · Executable planWrites the daily method for your restaurant's money in four named phases —desk opening, control during service…
Included in the always-growing library
Access to every published assistant, adapted to your AI and personalised with your restaurant's data, updates included.