📜 HumanizeIQ AI Development Rules

This document outlines the strict boundaries and guidelines for AI-driven modifications to this application. These rules ensure that the "Wiring" (Auth, API Proxies, RBAC) and "Branding" (Standard Layout) remain functional and consistent across all Aura Craft Studio projects.

🚫 1. Immutable "Wiring" (DO NOT MODIFY)

The following files and logic sections are core infrastructure. Modification may break connectivity with the HumanizeIQ backend or Google AI Studio.

services/apiUtils.ts: Logic for switching between Studio and Deployed modes.
services/geminiService.ts: The proxy-aware AI client initialization.
services/apiService.ts: R2 storage integration methods.
metadata.json: The application's DNA.
local_cookie.json: Essential for authentication within Google AI Studio.

🏗️ 2. Architecture: Container/View & SRP

To maintain scalability and the Single Responsibility Principle, follow these patterns:

Logic vs. Display Separation: Business logic, data fetching, and state orchestration must live in Containers (*Container.tsx) or Hooks (use*.ts). UI layout and styling must live in Views (*View.tsx or *Layout.tsx).
Max File Length: No code file should exceed 200 lines. If a file grows beyond this, it MUST be decomposed into smaller, more specific modules, sub-components, or custom hooks.
Modular Features: New capabilities should be grouped under features/[feature-name]/ with their own logic and display files.

✅ 3. Permitted Modification Areas

features/: Create domain-specific modules.
components/: Create shared UI atoms or molecules.
types.ts: Add new interfaces specific to the application's domain.
hooks/: Add custom React hooks for local state or specific data fetching.

🎨 4. Aesthetic & Theme Standards

Tailwind Strategy: Always use Tailwind CSS classes. Avoid inline styles.
Dark Mode: Every component must include dark: variants.
Responsiveness: Ensure layouts are mobile-friendly using sm:, md:, and lg: prefixes.

🤖 5. Integration Best Practices

Data Fetching: Use getUrlWithStudioAuth and getFetchOptions from apiUtils.ts.
AI Prompts: Use getSystemInstruction(key) from geminiService.ts.
Error Handling: Implement loading states and graceful fallbacks.

2.4 KiB Raw Blame History