Final Project & Overall Review

CLAUDE.md (Week 1)
  ├── Project context & conventions
  ├── Custom Commands (Week 4) ── reusable /slash prompts
  └── SKILL.md (Week 3) ── discoverable skill definitions

Hooks (Week 2)
  ├── PreToolUse ── block/modify before execution
  ├── PostToolUse ── validate after execution
  └── TypeScript checker, query duplication prevention

MCP Servers (Week 3-4)
  ├── Notion, Playwright, GitHub, PostgreSQL
  └── Multi-server pipelines for complex workflows

Subagents (Week 4)
  ├── Explore, Bash, Plan, General-Purpose
  ├── Parallel development with Task tool
  └── Background execution (run_in_background)

Headless Mode (Week 4)
  ├── claude -p "prompt" ── non-interactive execution
  ├── --output-format json ── structured output
  └── CI/CD pipeline integration (Week 5)

name: CI Pipeline

on:
  push:
    branches: [main]
  pull_request:
    branches: [main]

jobs:
  build-and-test:
    runs-on: ubuntu-latest
    steps:
      - uses: actions/checkout@v4
      - uses: actions/setup-node@v4
        with:
          node-version: '20'
      - run: npm ci
      - run: npm run build
      - run: npm test

name: Claude Code Review

on:
  pull_request:
    types: [opened, synchronize]

jobs:
  review:
    runs-on: ubuntu-latest
    steps:
      - uses: actions/checkout@v4

      - name: Install Claude Code
        run: npm install -g @anthropic-ai/claude-code

      - name: Review PR Changes
        env:
          ANTHROPIC_API_KEY: ${{ secrets.ANTHROPIC_API_KEY }}
        run: |
          claude -p "Review git diff for bugs, security issues,
            and code quality. Output as JSON." \
            --output-format json \
            --allowedTools "Read,Grep,Glob" \
            --yes > review-report.json

      - name: Upload Report
        uses: actions/upload-artifact@v4
        with:
          name: review-report
          path: review-report.json

name: Pre-merge Check

on:
  pull_request:

jobs:
  validate:
    runs-on: ubuntu-latest
    steps:
      - uses: actions/checkout@v4
      - uses: actions/setup-node@v4

      - run: npm ci
      - run: npm run lint
      - run: npm run build
      - run: npm test

      # Optional: Claude security scan
      - name: Security Scan
        env:
          ANTHROPIC_API_KEY: ${{ secrets.ANTHROPIC_API_KEY }}
        run: |
          npm install -g @anthropic-ai/claude-code
          claude -p "Scan for OWASP Top 10 vulnerabilities" \
            --allowedTools "Read,Grep,Glob" --yes

# 1. Initialize project with context (Week 1)
> /init
> # Add project conventions to CLAUDE.md

# 2. Plan the feature (Week 1)
> Shift+Tab x2 (Plan Mode)
> "Design user authentication system"

# 3. Implement with hooks for validation (Week 2)
# PostToolUse hook runs tsc after every file edit

# 4. Use subagents for parallel tasks (Week 4)
> "Build backend API while testing frontend components"

# 5. Deploy with CI/CD (Week 5)
> git push  # GitHub Actions auto-test & deploy

# Hook (Week 2): PostToolUse on Edit/Write
# Runs after every file modification

# 1. TypeScript validation hook
tsc --noEmit  # Catch type errors immediately

# 2. Query duplication hook
# Launches a review subagent (Week 4) to check for duplicates

# 3. CI/CD (Week 5): On push
# GitHub Actions runs full test suite + Claude security scan

# 4. Headless review (Week 4): On PR
claude -p "Review this PR" --output-format json --yes

# Step 1: Research with MCP (Week 3-4)
# Playwright MCP scrapes competitor data
# Notion MCP fetches product requirements

# Step 2: Parallel Development with Subagents (Week 4)
Task: Plan architecture
Task: Build API endpoints
Task: Create UI components
Task: Write tests

# Step 3: Quality with Hooks (Week 2)
# Auto-lint, type-check, security scan on every edit

# Step 4: Deploy with CI/CD (Week 5)
# Push to main -> GitHub Actions -> Vercel/Railway

my-project/
├── CLAUDE.md              # Project context & conventions
├── .claude/
│   ├── settings.local.json # Hooks + Custom Commands + MCP
│   └── skills/            # Custom skills (SKILL.md)
├── .github/
│   └── workflows/
│       ├── ci.yml         # Build + Test pipeline
│       └── review.yml     # Claude headless review
├── hooks/
│   ├── tsc_check.js       # PostToolUse: type checking
│   └── env_guard.js       # PreToolUse: block .env reads
├── src/                   # Application source code
├── tests/                 # Test suite
└── package.json

# Project Name

## Architecture
- Frontend: React + TypeScript
- Backend: Express.js + PostgreSQL
- Deploy: Vercel (frontend) + Railway (backend)

## Conventions
- Use functional components with hooks
- All API routes return JSON with {data, error} shape
- Tests required for all new features
- Never commit .env files

## Common Tasks
- Build: npm run build
- Test: npm test
- Lint: npm run lint
- Dev server: npm run dev