clone-deployment

---

name: clone-deployment description: Production deployment -- Docker, CI/CD, domain, monitoring, launch checklist category: clone-pipeline trigger: "deploy clone", "production deployment", "launch clone", "clone deployment"

Clone Deployment

When to Use

• Testing complete, quality verdict is PASS (output of clone-testing)
• Ready to ship to production
• Need repeatable, automated deployment pipeline

Process

1. Containerization:
   • Dockerfile per service (frontend, backend, worker)
   • docker-compose.yml for local orchestration
   • Multi-stage builds for minimal image size
   • Health check endpoints in each service
2. CI/CD pipeline (GitHub Actions):
   • On push: lint --> type-check --> unit tests --> build
   • On PR merge to main: full test suite --> build images --> push to registry
   • On release tag: deploy to production
   • Secrets managed via GitHub Secrets / Vault
3. Domain and SSL:
   • DNS A/CNAME records (Cloudflare or provider)
   • SSL via Let's Encrypt (auto-renewal)
   • Reverse proxy (nginx / Caddy)
4. Monitoring:
   • Uptime monitoring (UptimeRobot, Uptime Kuma)
   • Error tracking (Sentry)
   • Performance APM (response times, error rates)
   • AI cost tracking (tokens used, cost per request)
5. Launch checklist:
   • All ENV variables set in production
   • Database migrations applied
   • SSL certificate valid
   • Monitoring alerts configured
   • Backup strategy in place
   • Rate limiting enabled
   • CORS configured for production domain
6. Post-launch verification:
   • Smoke test critical path on production URL
   • Verify monitoring receives data
   • Check logs for errors in first 15 minutes

Output

• Deployed application accessible via production URL
• CI/CD pipeline (green builds)
• Monitoring dashboard with alerts
• Deployment runbook for future releases

Example

SlideDeck AI deployed: frontend on Vercel, backend on Railway (Docker), PostgreSQL on Supabase. CI/CD: GitHub Actions with 3 stages, 4 min total. Domain: slidedeck.example.com with Cloudflare DNS + SSL. Monitoring: Sentry for errors, UptimeRobot for uptime, custom dashboard for AI token spend ($0.03/generation).