AI assistant · N5 Complete system
Dashboard & KPI Agent for Restaurants
Not a one-shot prompt: an agent you install that works every week — takes your figures, updates the logbook, reads the lights and returns three alerts in money and one proposed decision.
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
- Dashboards die a silent death: built with enthusiasm, unfed and unread by week three.
- What's missing isn't the board — it's the analyst who takes the figures every Monday, compares them against target and says what they mean.
- This assistant installs that analyst: a standing agent with a defined mandate, inputs, limits and output format, keeping the logbook as memory (because chat forgets) and returning the same thing every week: lights read, three alerts stated in money and ONE proposed decision with two paths.
- And it declares what it does NOT decide: the agent proposes, the owner disposes.
What you end up with
An installed agent with a written mandate and limits; the KPI logbook serving as memory between weeks; and every Monday the same delivery: lights against target, three alerts in money with probable cause, one proposed decision with two paths and the missing figure — asked for, never invented.
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 installed the agent a month ago. Pasting week 5's close: sales 51.2M, food cost 34.8%, labor 27.5%, ticket 47,200, 62 deliveries.»
| Indicator | This week's value | Color | Ceiling or target | Weeks in that color |
|---|---|---|---|---|
| Sales | 51.2 M | 🟢 | weekly target 50 M | 2 |
| Food cost | 34.8% | 🔴 | ceiling declared by the house | 3 |
| Labor cost | 27.5% | 🟡 | ceiling declared by the house | 1 |
| Average ticket | 47,200 | 🟡 | target 48,000 | 4 |
| Deliveries | 62 | 🔴 | logbook's own mean: 81 | 2 |
THE THREE ALERTS, IN MONEY — ordered priciest first.
- Food cost, 1.43 M this week. Third week in red. The logbook shows the jump in week 3, which is the protein supplier change: that is the cause the series supports, not grammage, which did not change hands.
- Deliveries, −0.9 M/week at the current ticket. The drop also starts in week 3, when the platform moved its schedule.
- Average ticket, −0.4 M a month. Four weeks under target, always slightly. Not urgent; logged so it does not become habit.
SUPUESTO: the food cost ceiling is the one this house declared on its card, not a sector average. If the real ceiling were 2 points higher, this week would come out amber, the alert would drop to second place and the decision below would change indicator.
THE PROPOSED DECISION — one only: food cost, the priciest alert and the only one with a probable cause identified in the series.
| Path A · audit the supplier | Path B · audit portions | |
|---|---|---|
| What gets done | Request the price comparison against the previous supplier and re-cost the 5 highest-protein dishes | Weigh 10 served portions against the standard recipe for 1 week |
| Whose time it costs | One afternoon of the manager's | 2 hours of the head chef's |
| Figure that confirms it | Food cost ≤ 33% within 2 weeks | Grammage deviation under 5% |
| When it gets reread | Week 7 | Week 6 |
The agent does NOT choose: its mandate forbids it. Pick a path and the next delivery brings its plan.
THE MISSING FIGURE: week 5's food purchases never arrived. Without them the food cost stays estimated on the previous inventory and the alert loses its evidence. Asking the manager: what is the total purchase amount for week 5?
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 updated logbook**: the full series chat cannot remember, the week's row written and the open-decisions checklist with its verification figure and its reread date.
See the full example — free accountWorks with these AIs
How to use it
Which of your data it uses
This assistant works with 9 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.
v1.1.02026-08-25Body rewritten to pay down duelo debt: it scored 69.9 out of 90.0. The `cifras` signal read 1 of 12 —an agent that lives on weeks, streaks and reread dates carried almost none with a unit—, `lexico` 3 of 10 and `verificacion` 1 of 5; `metodo` was 4 of 6 because installation and the running week were mixed into one paragraph. The TASK is now six numbered steps —1 and 2 run once, 3 to 6 every week— injecting «fields», the fixed OUTPUT FORMAT sits in four blocks with their columns, and BEFORE YOU DELIVER carries five checks. The hard-coded «32% food cost» is removed from the method: the ceiling is declared by the house, and anything from outside carries the SUPUESTO line.v1.0.02026-08-15Initial version. Born in batch 3 of the artifact titles, opens the Specialized AI Agents category, with the full lead-magnet kit and a Sazón de Origen example. Its hallmark: defined mandate, inputs, limits and output format — the agent proposes, the owner disposes.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
You pick which agent to create —one of the MASTERESTAURANT firm's seven departments (Marketing, Commercial, Ma…
N5 · Complete systemThe table already exists and you can download it: twenty-three experts with deep profiles and fourteen guardia…
Included in the always-growing library
Access to every published assistant, adapted to your AI and personalised with your restaurant's data, updates included.
