AI Customer Service area · small businesses

Customers book, confirm and ask questions on WhatsApp. The team steps in only when needed.

Six sectors, one design: scheduling or booking, automatic confirmation, a waiting list, frequently asked questions, a handover to a human on Telegram, and health and privacy rules built in. Here you will find the system that already runs (clinic), five ready-made kits to implement in your own environment (restaurant, hotel, gym, physiotherapy and nutrition) and the method and tools used to build all of it with agents.

All open and free. Guides in PT, EN and ES. Dependency-free Python, Docker for a VPS, WhatsApp through the Evolution API and the team on Telegram.

01 · What already runs

Three projects ready to use today.

A complete clinic system, a framework that turns a website into an assistant, and an open CRM for WhatsApp.

02 · Kits by sector

Five ready-made plans: you run the /goal and the agent builds.

Each kit comes with the full specification, 93 to 99 frozen acceptance tests, the guardrails and the prompts for Claude Code or Codex to implement the system in your environment. The system itself is not built yet: it is the same path that took the clinic to 93 of 93.

🍽️ Restaurant · 93 tests

Table booking

Tables and rooms, shifts, the smallest table that fits, confirmation, no-shows, a waiting list, large groups with approval and your own customer base outside the marketplace.

Open the guide →
🏨 Hotel / inn · 98 tests

Direct booking

Per-night availability without overbooking, seasonal rates, a direct-booking discount, pre-check-in, iCal with Booking and Airbnb, and the source of each booking to measure the commission avoided.

Open the guide →
🏋️ Gym · 99 tests

Classes, workouts and animated exercises

Plans and enrollment, classes with spots and a waiting list, check-in, re-engaging members who disappeared, a workout sheet and an animated explanation of 12 exercises.

Open the guide →
🦵 Physiotherapy · 98 tests

Home exercises, adherence and pain

Treatment plan, a dropout alert, prescribed exercises with an animated demo, reminders, "done / not done / pain from 0 to 10" and an alert to the physiotherapist. COFFITO lock (Brazil's physiotherapy council).

Open the guide →
🥗 Nutrition · 99 tests

Meal plan and food diary

In-person or online consultation, a meal plan with versions, a food diary, water and weight with a neutral reply, and alerts that lead to 192 or to CVV (188), Brazil's emergency and crisis lines. The bot never prescribes.

Open the guide →
🩺 Clinic · implemented

The template for all of them

The first to go through the whole cycle: plan, frozen tests, long run, independent verification and publication.

Open the guide →
Animated demo. In the gym and physiotherapy kits, each exercise is a CSS-only animated SVG for the website and an MP4 generated by HyperFrames for WhatsApp, which doesn't display SVG. Validation measured in a real render that HyperFrames doesn't play SMIL animation, so the rule is "CSS only".

03 · The service flow

From the "hi" on WhatsApp to the monthly numbers.

WhatsApp or site chat→Frequently asked questions→Scheduling or booking→Confirmation and waiting list→Team on Telegram→Numbers for the X-Ray
The bot

Answers only what it knows

The company's own frequently asked questions, scheduling and booking. Anything else, it says it didn't understand and, the second time, calls a human. It never makes up a price, a diagnosis or a diet.

The rules

Built in, not remembered

A reminder is part of the duty of care in health; a campaign requires consent; "STOP" blocks everything. The campaign text goes through the sector's professional-council lock before it goes out. The built-in rules are Brazilian (LGPD, CFM, CFO, COFFITO, CFN, CREF).

The team

In the Telegram group

Receives each conversation with a number, replies on top of the message and uses commands such as /fila, /agenda and /encerrar. An emergency arrives flagged as HIGH.

Automated service doesn't replace the front desk: it takes away what is repetitive and gives back what needs a person.

04 · How they were built

Agents plan, other agents attack the plan, and the test decides.

Each kit was written by one agent (Fable) and validated by another that hadn't seen the planning. The validators found 9 to 13 problems per kit, 1 to 3 of which would have blocked execution, and fixed all of them before publication. Implementation runs through a long run, and the runtime is what lets the agent use its tools safely.

What validation caught. A test that demanded the opposite of the specification, a test server that kept running when the test failed, a container that wouldn't start without a volume, a verifier that reported a false failure on this machine, and a hash that didn't protect the verifier itself. All of it became a rule in the template.

05 · In your environment

Four steps from kit to a live system.

1 · Decide

Answer the decisions

Each kit has docs/DECISOES-ABERTAS.md with the default proposal already applied. "OK on everything" is enough to start.

2 · Freeze

Lock the contract

bash longrun/<folder>/congelar.sh records the hash of the specification, the tests and the verifier.

3 · Run

Loop or /goal

The headless loop from execucao-longa (Codex through the subscription) or /goal in Claude Code or Codex. It ends only when every test passes.

4 · Publish

Docker on a VPS

docker compose up -d, an HTTPS proxy, the webhook for your Evolution and your Telegram bot. Credentials stay only in your .env.

Honest caveats. The exercises and the alert lists are examples and need review by a professional. The rules of CFN (nutrition) and CREF (physical education), Brazilian professional councils, have not been researched yet: the campaign lock in those kits is minimal until the owner confirms.

06 · Why these systems

Each piece attacks a leak measured by the Margin X-Ray.

Clinic, physio and nutrition

No-shows without notice

The clinic's leak no. 1: 5 to 12% in a well-run schedule, close to 25% on average. Confirmation, a waiting list and scheduled follow-ups attack it directly.

Restaurant

A customer who is yours, charged again

The marketplace charges a commission on someone who is already the house's customer. Your own booking and a consented customer base win the channel back.

Hotel

The OTA's cut

Commission on repeat guests and empty rooms in the low season. Direct booking, an off-season package and iCal with Booking and Airbnb.

The systems export the month's numbers (no-shows, occupancy, new customers, returns) in the format of the Recovery Dashboard: the X-Ray's "before" and the "after" measured on the real schedule. See the Margin X-Ray area →

07 · Projects and courses

Everything from INEMA to build and run AI customer service.

Projects and kits

Courses

08 · Related areas

Other Events areas for people who deploy agents in businesses.

09 · Get started

The best first step: open the kit for your sector.

Read the open decisions, adapt the example to a business you know and run the /goal. When the agent says "done", run the independent verification.

System · implemented

Atende Clínica

The template for every kit, with 93 of 93 tests and Docker.

Open the guide →
Kits · five sectors

Pick yours

Restaurant, hotel, gym, physiotherapy or nutrition.

See the kits ↑
Method

Long Runs

How to let the agent build the system for hours, with proof at the end.

Open the guide →

Not sure where to start? Talk to us on Telegram.