Mise Kitchen Recipe Manager
The original prompt, shown in full as plain text.
Prompt
# Design & Architecture
## ๐ Tech Stack
> It may be necessary temporarily to accept a lesser evil, but one must never label a necessary evil as good.
* **Framework:** Next.js 16 (App Router + Server Actions).
* **UI:** Ionic React (Mobile behavior/navigation).
* **Styling:** @emotion/css (Simple toolchain for good old CSS... in JS).
* **Database:** Prisma + PostgreSQL.
## ๐ Key Decisions
* **Tech Stack:** Fastest, highest-fidelity stack possible with maximum ROI on developer experience.
* **No Tailwind:** Explicitly avoided to prevent utility-class bloat and maintain a precise, platform-native design system via standard CSS-in-JS.
* **Tenant Isolation:** Enforced at the query level; every record is hard-linked to an `Organization`.
* **Kitchen-First UI:**
* BIG typography (64px titles) for station-to-station readability on overhead tabled displays, but still compatible with handheld mobile.
* Large touch targets (~48px) for fast-paced kitchen use.
* High-contrast color palette (#030213 on White) for brightly-lit kitchens.
## ๐งช Quality Gates
* **Vitest:** "Business logic" and component unit tests. Basically just dummy tests to show how to do the thing.
* **Playwright:** E2E flows (Wizard, Search, Kitchen View). These simulate actual user workflows and include setup/teardown lifecycle.
* **Bonus:** reuse E2E tests with Artillery framework for load/scalability testing if necessary in future. Workflow/Scenario-based E2E tests help to create scalability tests that approximate real user behaviors.
## Alternatives Considered
- Rust + Godot game engine (for fun!)
- Golang backend + protobufs as single source of data model truth (for fun!)
- AWS RDS and Vanilla SQL (for business!)
Chosen stack was selected for a variety of reasons including rapid turnaround time. In production, might make some different choices altogethe after doing a proper trade study.