← Back to library

Prototype a booking app for a solo service business

Ilias Hatzis, a former programmer a decade away from serious coding, prototyped booking software for tattoo artists over a weekend using Replit Agent. A single descriptive paragraph produced scheduling, Google sign-in, a Stripe connection and Postmark notification emails, followed by a long tussle over time zones and polish. Artists get dashboards, availability controls and onboarding.

  • OutcomeBuild a working booking flow, with availability, confirmations and an artist dashboard, over a weekend without a dev team.
  • SetupAn afternoon+
  • LevelIntermediate
  • CreditIlias Hatzis
View original
Works with
Custom agent

Published Oct 23, 2026Updated Oct 23, 2026AI dry run on sample input: passed Oct 23, 2026Source verified Oct 23, 2026

When should you use it?

  • Clients book you through scattered DMs and no-shows are common
  • You want to test an idea before paying a developer
  • You can review an app's flows carefully

Not for: Taking real payments before security and refunds are reviewed; Launching without terms, privacy notice and testing

What do you need, and what do you get?

You need

  • A one-paragraph product description
  • Accounts for payments, email and sign-in
  • Sample services and availability

You get

  • A page where clients pick a slot
  • An artist dashboard for bookings and availability
  • Confirmation emails

How does it work, step by step?

  1. Describe the product in one paragraphName the calendar style, sign-in, payments and email services you want.
  2. Let it scaffold everythingAccept the backend, integrations and front end it generates first.
  3. Push on the hard partsTest time zones and double bookings yourself; that is where it struggled.
  4. Watch for regressionsIt sometimes rewrote working code; retest flows after each change.
  5. Hold payments until reviewedKeep deposits switched off until someone checks security and refunds.

The prompt

Our own prompt, inspired by Ilias Hatzis and written to work in Custom agent. Values in lime are placeholders.

Prompt0 variables
Create appointment software for a one-person [service] business. Clients see the provider's availability, choose a service and time, and receive a confirmation email. Providers sign in, set availability and services, and see a dashboard of bookings. Use [sign-in, email and payment services], but keep payments in test mode. Handle time zones explicitly and block double bookings. After each change, list which flows I should retest.

Inspired by Ilias Hatzis: From Zero to Working Prototype in 48 Hours with Replit. The prompt above is written by playbooks.bot in our own words; nothing from the original is quoted.

What can go wrong?

  • The agent occasionally said a bug was fixed when it was not; verify by testing.
  • Time-zone logic needed repeated correction.

Our take

Ilias Hatzis shows how quickly a niche product idea can now become something clickable, and also where that speed runs out. One paragraph got a working skeleton; the hard parts, time zones and polish, took real effort, and the post honestly records the agent rewriting working code and claiming fixes it had not made. Treat this as a way to validate an idea with real users, not as a launch. Before taking bookings, test the scenarios that hurt small businesses: overlapping appointments, a client in another time zone, daylight saving changes, cancellations and no-shows. Decide deposit and refund rules in writing, because software only enforces what you define. Keep payments in test mode until someone with security experience has reviewed authentication, password resets and how client data is stored, since a booking app holds names, phone numbers and sometimes health notes. Commit often so regressions can be undone, and keep a short log of what you asked for and what changed. Show it to two or three artists before building more.

Where the time goes back: A product idea becomes a clickable, testable prototype in a weekend instead of a hiring decision.

FAQ

Is it launched?
No. The author describes it as a functional prototype that had not yet launched.
Were payments live?
Stripe was connected and configured, with deposit handling on the roadmap.

Source and credit

Inspired by Ilias Hatzis. We describe the idea in our own words, link to the original, and wrote the prompt ourselves; nothing is quoted.

Inspired by
From Zero to Working Prototype in 48 Hours with Replit
Creator
Ilias Hatzis
Published on
iliashatzis.substack.com (Substack)
Original date
Jul 7, 2025
License
Not openly licensed (the poster keeps all rights), so we quote nothing and link to the original
Originally for
Custom agent
AI dry run
Oct 23, 2026: passed. Dry-run tested by our AI test agent on sample input. PASS: Gates held on desk check; no live sends or purchases.
AI dry run on sample data; not run with live connected accounts. The playbooks.bot test agent filled the prompt with synthetic inputs, produced the output and checked it against the page's stated outputs and guardrails.
Verified
Oct 23, 2026: source link resolves to the original, the credited creator published it, our excerpt or prompt matches it, and the license allows our use.

Are you the creator? Claim or correct this page.

More in Operations & Admin

All 203 Operations & Admin playbooks →

Related playbooks

Operations & AdminPrompt✓ AI dry run

Audit how you manage your agent team, not only the agents

Have your chief-of-staff agent review the last 30 days of how you delegated, clarified, interrupted, corrected, assigned, approved and followed up across your agent team, score eight management areas with evidence, list your three biggest weaknesses, and draft a seven-day improvement plan, without changing any bot, routine or instructions until you approve.

Credit: Michael Fenech (@Michael_Fenech_)

Under an hour
Operations & AdminPrompt✓ AI dry run

Audit how your agents pick tools before buying more usage

Have your lead agent review the recurring jobs it can see, note which apps, scripts, browser sessions or full computer control each one relies on, flag jobs that take a heavier route than needed or rely on remembered facts, check permissions and fallbacks, and then propose three side-by-side experiments before anything is changed.

Credit: Michael Fenech (@Michael_Fenech_)

Under an hour
Operations & AdminPrompt✓ AI dry run

Brief an agent with nine fields, not a how-to

Before you hand a job to an agent, fill in nine short fields that name the result, the sources it may use, the deadline, what it may do alone, what needs your say-so, what it must never do, how to prove the work is finished, what to do when it gets stuck, and where to put the deliverable, then add a line that tells it to pick the simplest reliable method.

Credit: Michael Fenech (@Michael_Fenech_)

Minutes
Operations & AdminWorkflow✓ AI dry run

Map and trim workflows before adding AI

Before building AI into a company, map each workflow with the people who do it, cut needless steps and plan the system, then add automation on top.

Credit: Luke Pierce (@lukepierceops)

An afternoon+
Operations & AdminPrompt✓ AI dry run

Review your bot's standing instructions in five parts

Have one of your bots show you the instructions it follows on every job, point out what looks stale and why, then rebuild them with you, one question at a time, around its goal, what it owns, its hard limits, this month's priorities and where the detailed files live, ending with a revised version for your approval.

Credit: Michael Fenech (@Michael_Fenech_)

Minutes