Code Review SaaS - Documentation

AI-powered code review platform with public and authenticated modes.

📚 Documentation Index

Getting Started

• Architecture Overview - System design, technology stack, architecture decisions
• Implementation Guide - Step-by-step setup and development guide
• Database Schema - Complete database design and migrations

Technical Documentation

• AI Services - AI provider integration and service architecture (to be created)
• Frontend Components - Vue components and UI patterns (to be created)
• API Documentation - REST API reference for integrations (Phase 3)

Guides

• Docker Setup - Docker configuration and commands
• Makefile Commands - Development workflow commands

🚀 Quick Start

# 1. Setup environment
make setup

# 2. Configure AI provider
cp .env.example .env
# Add your DIGITALOCEAN_AI_API_KEY

# 3. Start development
make dev

# 4. Visit application
open http://localhost:8080

📖 Project Structure

ai-companion/
├── app/
│   ├── Models/              # Eloquent models
│   ├── Services/            # Business logic
│   │   ├── AI/             # AI provider integration
│   │   ├── CodeReview/     # Review processing
│   │   └── Analytics/      # Usage tracking
│   ├── Jobs/               # Queue jobs
│   ├── Http/Controllers/   # Route handlers
│   └── Actions/            # Reusable actions
├── database/
│   └── migrations/         # Database schema
├── resources/
│   └── js/
│       ├── components/     # Vue components
│       │   ├── review/    # Code review UI
│       │   ├── project/   # Project management
│       │   └── ui/        # shadcn/ui components
│       └── pages/         # Inertia pages
├── routes/
│   ├── web.php            # Web routes
│   └── api.php            # API routes
├── config/
│   └── ai.php             # AI configuration
└── docs/                  # Documentation (this folder)

🎯 Features

Public Mode (MVP - Phase 1)

• ✅ Anonymous code review submission
• ✅ Real-time token/cost estimation
• ✅ AI-powered analysis
• ✅ Inline comments with severity levels
• ✅ Security vulnerability detection
• ✅ Shareable review URLs (7-day expiry)
• ✅ Rate limiting (5/hour, 10/day)

Authenticated Mode (Phase 2)

• 🔲 User registration & authentication
• 🔲 Project management (create, organize reviews)
• 🔲 Permanent review storage
• 🔲 Review history with search
• 🔲 Project dashboard with statistics
• 🔲 Higher rate limits (50/hour, 200/day)

API & Integration (Phase 3)

• 🔲 REST API for CI/CD
• 🔲 API token management
• 🔲 GitHub Action integration
• 🔲 Webhook support

Future Enhancements (Phase 4+)

• 🔲 Multiple AI providers (OpenAI, Claude, Gemini)
• 🔲 Code suggestions with diff view
• 🔲 Export reviews (PDF, Markdown)
• 🔲 Advanced analytics
• 🔲 Team collaboration
• 🔲 Payment integration

🛠 Technology Stack

• Backend: Laravel 12, PHP 8.5
• Frontend: Vue 3, Inertia.js, TypeScript
• UI: shadcn/ui, Tailwind CSS 4
• Database: PostgreSQL 18
• Cache/Queue: Redis, Horizon
• Authentication: Laravel Fortify (session + 2FA)
• AI: Flexible provider system (DigitalOcean, OpenAI, Claude, Gemini)

📋 Implementation Phases

Phase	Timeline	Status	Description
Phase 1	Weeks 1-3	🟡 In Progress	MVP: Public anonymous reviews
Phase 2	Weeks 4-5	⚪ Planned	Authentication & projects
Phase 3	Weeks 6-7	⚪ Planned	API & GitHub integration
Phase 4	Week 8+	⚪ Future	Enhancements & scaling

🔐 Security

• Rate Limiting: IP-based for public, user-based for authenticated
• Code Sanitization: Input validation, length limits
• XSS Prevention: Vue auto-escapes output
• CSRF Protection: Laravel default tokens
• SQL Injection: Eloquent ORM prevents attacks
• API Authentication: Laravel Sanctum token-based

🧪 Testing

# Run all tests
make test

# Unit tests only
make test --testsuite=Unit

# Feature tests only
make test --testsuite=Feature

# Specific test
make test --filter=PublicReviewTest

# Code formatting
make pint

# Code style check
make pint-test

📊 Monitoring

Horizon Dashboard

http://localhost:8082/horizon

• Queue status
• Job throughput
• Failed jobs
• Metrics

Key Metrics

• Average review processing time
• Token usage per review
• Cost per review
• Error rates by AI provider
• Queue wait times

🤝 Contributing

Development Workflow

1. Create feature branch
2. Implement changes
3. Write tests
4. Run code formatter: make pint
5. Run tests: make test
6. Create pull request

Code Style

• Follow PSR-12 for PHP
• Use Laravel Pint for formatting
• TypeScript strict mode enabled
• Vue 3 Composition API preferred

📝 Environment Variables

See .env.example for all available options.

Required:

DIGITALOCEAN_AI_API_KEY=your_api_key
DIGITALOCEAN_AI_BASE_URL=https://api.digitalocean.com/v2/ai
DIGITALOCEAN_AI_MODEL=gpt-oss-120b

Optional:

AI_MAX_CODE_LENGTH=50000
AI_MAX_TOKENS=50000
PUBLIC_REVIEW_EXPIRY_DAYS=7

🐛 Troubleshooting

Queue not processing

make horizon-status
make queue-restart
make logs-horizon

AI provider errors

# Check logs
docker exec -it ai-companion tail -f storage/logs/laravel.log

# Test connection
docker exec -it ai-companion php artisan tinker
>>> app(\App\Services\AI\AIProviderManager::class)->driver()->isAvailable()

Database issues

make migrate-fresh
make db-cli

📞 Support

• Issues: GitHub Issues
• Discussions: GitHub Discussions
• Documentation: /docs

📄 License

This project is open-sourced software licensed under the MIT license.

---

Documentation Version: 1.0 (MVP Phase) Last Updated: 2026-02-15 Status: Under Active Development