Senior Frontend Engineer

Overview

Deliver production-grade frontend code following a structured three-phase workflow: context discovery, development, and handoff. This skill enforces strict quality standards including atomic design component architecture, comprehensive state management patterns, SSR/SSG/ISR optimization, and mandatory >85% test coverage with Vitest, React Testing Library, and Playwright.

Announce at start: "I'm using the senior-frontend skill for production-grade React/TypeScript development."

---

Phase 1: Context Discovery

Goal: Understand the existing codebase before writing any code.

Actions

1. Analyze existing codebase structure and conventions
2. Identify the tech stack version (React 18/19, Next.js 14/15, TypeScript version)
3. Review existing component library and design system
4. Check state management approach already in use
5. Understand build tooling and CI pipeline
6. Map existing test infrastructure and coverage

STOP — Do NOT proceed to Phase 2 until:

• Tech stack versions are identified
• Existing patterns and conventions are documented
• Test infrastructure is mapped
• State management approach is identified

---

Phase 2: Development

Goal: Implement with strict TypeScript, atomic design, and TDD.

Actions

1. Design component architecture following atomic design
2. Implement with TypeScript strict mode
3. Write tests alongside implementation (TDD when appropriate)
4. Optimize for performance (bundle size, rendering, loading)
5. Ensure accessibility compliance

Component Architecture Decision Table (Atomic Design)

Level	Description	Business Logic	Example
Atoms	Smallest building blocks	None	Button, Input, Icon, Badge
Molecules	Composed of atoms	Minimal	FormField, SearchBar, Card
Organisms	Complex with business logic	Yes	DataTable, NavigationBar, CommentThread
Templates	Page structure without data	Layout only	DashboardLayout, AuthLayout
Pages	Templates connected to data	Data fetching	UsersPage, SettingsPage

Atom Example

interface ButtonProps extends React.ButtonHTMLAttributes<HTMLButtonElement> {
  variant?: 'primary' | 'secondary' | 'ghost' | 'danger';
  size?: 'sm' | 'md' | 'lg';
  isLoading?: boolean;
}

export function Button({ variant = 'primary', size = 'md', isLoading, children, ...props }: ButtonProps) {
  return (
    <button className={cn(buttonVariants({ variant, size }))} disabled={isLoading || props.disabled} {...props}>
      {isLoading ? <Spinner size={size} /> : children}
    </button>
  );
}

State Management Decision Table

State Type	Solution	When to Use
Server state	React Query / TanStack Query	API data, caching, sync
Form state	React Hook Form + Zod	Form validation, submission
Global UI state	Zustand	Theme, sidebar open, modals
Local UI state	useState / useReducer	Component-specific state
URL state	nuqs / useSearchParams	Filters, pagination, tabs
Complex local	useReducer	Multiple related state transitions
Shared context	React Context	Theme, locale, auth (infrequent updates)

SSR / SSG / ISR Decision Table (Next.js App Router)

Pattern	Use When	Cache Strategy
Static (SSG)	Content rarely changes	Build time
ISR	Content changes periodically	Revalidate interval
SSR	Content changes per request	No cache
Client	User-specific, interactive	Browser

Server vs Client Component Decision

Need	Component Type
Direct data fetching	Server (default)
Event handlers (onClick, onChange)	Client ('use client')
useState / useReducer	Client
useEffect / useLayoutEffect	Client
Browser APIs (window, localStorage)	Client
Third-party libs using client features	Client
No interactivity needed	Server (default)

STOP — Do NOT proceed to Phase 3 until:

• Components follow atomic design hierarchy
• TypeScript strict mode is enabled, no any types
• Tests are written for all components
• Accessibility is verified (axe-core)

---

Phase 3: Handoff

Goal: Verify quality gates and prepare for review.

Actions

1. Verify test coverage meets >85% threshold
2. Run full lint and type check
3. Document complex components with JSDoc/TSDoc
4. Create Storybook stories for UI components
5. Performance audit (Lighthouse, bundle analysis)

Performance Checklist

• Bundle size < 200KB gzipped (initial load)
• Largest Contentful Paint < 2.5s
• First Input Delay < 100ms
• Cumulative Layout Shift < 0.1
• Images: next/image with proper sizing and formats
• Fonts: next/font with display swap
• No layout thrashing (batch DOM reads/writes)
• Virtualization for lists > 100 items

Coverage Thresholds

{
  "coverageThreshold": {
    "global": {
      "branches": 85,
      "functions": 85,
      "lines": 85,
      "statements": 85
    }
  }
}

STOP — Handoff complete when:

• Test coverage >85% verified
• Lint and type check pass with zero errors
• Performance audit completed
• Complex components documented

---

Testing Requirements

Unit Tests (Vitest + React Testing Library)

describe('Button', () => {
  it('renders children', () => {
    render(<Button>Click me</Button>);
    expect(screen.getByRole('button', { name: 'Click me' })).toBeInTheDocument();
  });

  it('shows loading state', () => {
    render(<Button isLoading>Click me</Button>);
    expect(screen.getByRole('button')).toBeDisabled();
  });

  it('calls onClick when clicked', async () => {
    const onClick = vi.fn();
    render(<Button onClick={onClick}>Click me</Button>);
    await userEvent.click(screen.getByRole('button'));
    expect(onClick).toHaveBeenCalledOnce();
  });
});

Integration Tests

• Component compositions (form submission flow)
• Data fetching with MSW (Mock Service Worker)
• Routing and navigation
• Error boundaries and fallbacks

E2E Tests (Playwright)

test('user can complete checkout', async ({ page }) => {
  await page.goto('/products');
  await page.getByRole('button', { name: 'Add to cart' }).first().click();
  await page.getByRole('link', { name: 'Cart' }).click();
  await expect(page.getByText('1 item')).toBeVisible();
  await page.getByRole('button', { name: 'Checkout' }).click();
});

---

React Query Patterns

function useUsers(filters: UserFilters) {
  return useQuery({
    queryKey: ['users', filters],
    queryFn: () => fetchUsers(filters),
    staleTime: 5 * 60 * 1000,
    placeholderData: keepPreviousData,
  });
}

function useUpdateUser() {
  const queryClient = useQueryClient();
  return useMutation({
    mutationFn: updateUser,
    onMutate: async (newUser) => {
      await queryClient.cancelQueries({ queryKey: ['users'] });
      const previous = queryClient.getQueryData(['users']);
      queryClient.setQueryData(['users'], (old) =>
        old.map(u => u.id === newUser.id ? { ...u, ...newUser } : u)
      );
      return { previous };
    },
    onError: (err, newUser, context) => {
      queryClient.setQueryData(['users'], context.previous);
    },
    onSettled: () => {
      queryClient.invalidateQueries({ queryKey: ['users'] });
    },
  });
}

---

Memoization Decision Table

Technique	Use When	Do NOT Use When
useMemo	Expensive computation, referential equality for deps	Simple calculations, primitive values
useCallback	Functions passed to memoized children	Functions not passed as props
React.memo	Component re-renders often with same props	Props change on every render
None	Default — do not memoize	Always profile first

---

Anti-Patterns / Common Mistakes

Anti-Pattern	Why It Is Wrong	Correct Approach
useEffect for data fetching	Race conditions, no caching, no dedup	React Query or Server Components
Prop drilling more than 2 levels	Tight coupling, maintenance burden	Composition, context, or Zustand
Business logic in components	Untestable, unreusable	Extract to hooks or utility functions
Barrel exports	Breaks tree-shaking, slower builds	Direct imports
Testing implementation details	Brittle tests that break on refactor	Test behavior: user actions and outcomes
any type anywhere	Defeats TypeScript's purpose	unknown + type guards
Inline styles for non-dynamic values	Inconsistent, hard to maintain	CSS modules, Tailwind, or styled-components
Memoizing everything	Adds complexity, often slower	Profile first, memoize second

---

Documentation Lookup (Context7)

Use mcp__context7__resolve-library-id then mcp__context7__query-docs for up-to-date docs. Returned docs override memorized knowledge.

• react — when uncertain about hooks API, component lifecycle, or React 19+ features
• next.js — for App Router, Server Components, or Next.js-specific APIs
• typescript — for advanced type patterns or compiler options
• tailwindcss — for utility classes, configuration, or plugin API
• vitest — for test runner API, matchers, or mock utilities

---

Integration Points

Skill	Relationship
testing-strategy	Strategy defines frontend test frameworks
test-driven-development	Components are built with TDD cycle
react-best-practices	Detailed React patterns complement this skill
performance-optimization	Frontend performance follows optimization methodology
code-review	Review verifies component architecture and test coverage
clean-code	Code quality principles apply to component code
webapp-testing	Playwright E2E tests use this skill's page structure
acceptance-testing	UI acceptance criteria drive component tests

---

Key Principles

• TypeScript strict mode, no any (use unknown + type guards)
• Prefer composition over inheritance
• Colocate tests, styles, and stories with components
• Server Components by default; Client Components only when required
• Error boundaries at route and feature boundaries
• Accessibility is not optional (test with axe-core)

---

Skill Type

FLEXIBLE — Adapt component architecture and state management to the existing project conventions. The three-phase workflow is strongly recommended. Test coverage must target >85%. TypeScript strict mode is non-negotiable.