name: architecture-philosophy description: Core architectural principles for StepLeague - modular design, system thinking, future-proofing, and maintenance reduction. Use when designing new features, refactoring code, considering implementation approaches, or making any architectural decisions. Keywords: architecture, design, modular, refactoring, reusable, shadcn, maintenance, future-proof, system. compatibility: Antigravity, Claude Code, Cursor metadata: version: "1.1" project: "stepleague"

Architecture Philosophy Skill

THE MOST IMPORTANT SKILL: This skill defines HOW we build everything. Every other skill references these principles.

⚠️ Critical: Read This First

Before implementing ANY solution:

1. Think systems, not solutions - Will this pattern be needed elsewhere?
2. Check existing patterns - Is this already solved in the codebase?
3. Consider maintenance - How many places need updating when requirements change?
4. Reference other skills - They encode project-specific implementations of these principles

Core Principles

Every solution in StepLeague must follow these principles:

1. 🧩 Modular Over Monolithic (MOST CRITICAL)

Build systems, not one-off solutions.

If you're writing code that solves just ONE specific case, you're probably doing it wrong. Take 5 extra minutes to think: "How can I make this reusable?"

Questions to ask BEFORE writing code:

• Will this pattern be needed elsewhere? → Extract it
• Can I extend an existing utility/hook/component? → Extend, don't duplicate
• Does this belong in a shared location? → Move it to /lib/ or /components/ui/
• Am I hardcoding a value that might change? → Use settings

2. 🔮 Future-Thinking

Design for tomorrow, implement for today.

• Settings over hardcoding: Use app_settings table for configurable values
• Feature flags: Gate new features with useFeatureFlag()
• Extensible schemas: Add columns that allow future use cases

// ❌ WRONG: Hardcoded limit
const MAX_PROXIES = 10;

// ✅ CORRECT: Configurable
const maxProxies = useAppSettings('proxy_max_per_user', 10);

3. 🛡️ Defensive Programming

Assume things will fail. Plan for it.

• Use the error handling system - Reference error-handling skill
• Graceful degradation - If a feature fails, don't crash the page
• Fallback values - Always have sensible defaults

// ✅ Defensive pattern with fallback
const branding = await getCachedBranding().catch(() => DEFAULT_BRANDING);

4. 📉 Maintenance Reduction

Less code = less bugs = less maintenance.

Mandatory Patterns

A. Use shadcn/ui Components

ALWAYS prefer shadcn over custom implementations.

// ✅ Use shadcn
import { Dialog } from "@/components/ui/dialog";
import { DropdownMenu } from "@/components/ui/dropdown-menu";
import { toast } from "@/hooks/use-toast";

// ❌ Don't create custom modals, dropdowns, or alerts

Installed shadcn components:

• toast, dialog, dropdown-menu, input, label
• textarea, select, checkbox, tooltip
• confirm-dialog (custom wrapper)

Reference: src/components/ui/

B. Use Reusable Form Components

Reference the form-components skill.

// ✅ Use form fields
import { FormInput, FormSelect, FormCheckbox } from "@/components/ui/form-fields";

// ❌ Don't use raw <input>, <select> elements

C. Use the API Handler Pattern

Reference the api-handler skill.

// ✅ Use withApiHandler
export const POST = withApiHandler({
  auth: 'required',
  schema: mySchema,
}, async ({ user, body, adminClient }) => {
  return { success: true };
});

// ❌ Don't write boilerplate auth checks in every route

D. Use the Error System

Reference the error-handling skill.

// ✅ Use AppError
throw new AppError({
  code: ErrorCode.VALIDATION_FAILED,
  message: 'Invalid input',
  context: { field: 'email' },
  recoverable: true,
});

// ❌ Don't throw generic Error objects

E. Use CSS Variables for Theming

Reference the design-system skill.

// ✅ Semantic variables
<div className="bg-card text-foreground border-border">

// ❌ Hardcoded Tailwind colors
<div className="bg-slate-900 text-slate-300 border-slate-700">

Decision Framework

When implementing any feature, follow this checklist:

Before Writing Code

1. Check existing patterns in AGENTS.md
2. Search for similar implementations in the codebase
3. Review relevant skills in .agent/skills/
4. Consider: Is this a one-off or a pattern?

While Writing Code

1. Extract reusable pieces into utilities/components
2. Use centralized configuration (settings, registries)
3. Add proper error handling with AppError
4. Follow the design system for all UI

After Writing Code

1. Update documentation if adding new patterns
2. Add to design system page if creating UI components
3. Reference in AGENTS.md if it's a key pattern
4. Create a skill if it's complex enough

Cross-Skill References

Anti-Patterns to Avoid

1. The "Quick Fix"

"I'll just add this inline, it's faster"

Why it's bad: Creates tech debt, makes code harder to maintain.

Instead: Take 5 extra minutes to do it properly.

2. The "Special Case"

"This component is unique, it needs custom styling"

Why it's bad: Breaks consistency, harder to theme.

Instead: Use existing components, extend if needed.

3. The "Magic Number"

if (count > 50) { ... }

Why it's bad: Unexplained, not configurable.

Instead: Use settings or named constants.

4. The "Silenced Error"

try { ... } catch { /* ignore */ }

Why it's bad: Hides bugs, makes debugging impossible.

Instead: Use normalizeError() and reportErrorClient().

5. The "Copy-Paste Component"

Creating a new component that's 90% the same as existing one.

Why it's bad: Duplicated code, inconsistent behavior.

Instead: Extract shared logic, use props for variations.

The 3-Time Rule

If you do something 3 times, extract it.

Key Reference Files

Summary

Build systems, not solutions.

Every feature you implement should make the next feature easier to build.