AI Coding Assistants

# Architecture review
"Review this Next.js API route for security vulnerabilities,
performance issues, and TypeScript best practices. Suggest
specific improvements with code examples."

# Code generation with constraints
"Write a custom React hook called useDebounce that:
- Takes a value and delay in milliseconds
- Returns the debounced value
- Properly cleans up timers on unmount
- Has full TypeScript generics support
- Includes JSDoc documentation"

# Debugging assistance
"This React component re-renders on every keystroke even
though I'm using React.memo. Here's the component and its
parent. Explain why memoization is failing and show the fix."

# Code migration
"Convert this class component to a functional component with
hooks. Preserve all behavior including lifecycle methods,
error boundaries, and ref forwarding."

# Cursor keyboard shortcuts
Cmd+K          # Inline edit (describe what to change)
Cmd+L          # Open AI chat
Cmd+Shift+L    # Add selection to chat context
Cmd+I          # Composer (multi-file editing)

# Cursor features:
# - Codebase indexing: indexes your entire project for context
# - @-mentions: reference files, functions, or docs in prompts
# - Multi-file edits: AI can modify multiple files in one action
# - Tab completion: predictive edits based on recent changes
# - .cursorrules: project-specific AI instructions

# .cursorrules - Project configuration for Cursor
You are working on a Next.js 15 project with:
- App Router (not Pages Router)
- TypeScript strict mode
- Tailwind CSS for styling
- Prisma ORM with PostgreSQL
- Vitest for testing

Conventions:
- Use server components by default
- Use 'use client' directive only when state or browser APIs are needed
- Prefer Server Actions over API routes for mutations
- Use Zod schemas for validation
- Follow the existing code patterns in the codebase

// Recommended AI workflow for a new feature:

// 1. PLAN: Use Claude/ChatGPT chat to discuss architecture
//    "I need to add a notification system. Here's my current
//     architecture. What's the best approach?"

// 2. SCAFFOLD: Use Copilot/Cursor to generate boilerplate
//    Write a descriptive comment and let AI generate the structure

// 3. IMPLEMENT: Use inline completion for routine code
//    Focus on the complex logic yourself, let AI handle the rest

// 4. TEST: Use /tests command to generate test scaffolds
//    Review and adjust the generated tests

// 5. REVIEW: Ask AI to review your implementation
//    "Review this PR for bugs, security issues, and performance"

// 6. DOCUMENT: Use /doc to generate documentation
//    Edit the generated docs for accuracy and clarity