Deployment Skill

Pre-Deployment Checklist

Before deploying to production:

Code Quality

• • All tests passing (unit, integration, e2e)
• • Code coverage ≥80%
• • No type errors
• • No linter errors or warnings
• • No formatting issues
• • Security vulnerabilities checked

Build & Configuration

• • Production build succeeds locally
• • Environment variables documented in .env.example
• • All required secrets configured
• • Database migrations ready and tested
• • Migration rollback scripts prepared

Deployment Plan

• • Rollback procedure documented
• • Stakeholders notified
• • Monitoring and health checks configured
• • Post-deployment verification steps defined

Environment Variables

Create .env.example:

# Document all required variables
DATABASE_URL=
API_KEY=
JWT_SECRET=
NODE_ENV=production

Runtime Validation: Validate environment variables at application startup to catch configuration errors early.

Database Migrations

Safe Migration Strategy:

1. •Test migrations in staging environment
2. •Backup production database before migration
3. •Write rollback scripts for all migrations
4. •Apply migrations before or after code deployment (depending on breaking changes)
5. •Monitor application after migration

Health Checks

Implement health check endpoints:

• •/health - Overall system health (database, cache, external services)
• •/ready - Readiness to serve traffic
• •/alive - Process liveness

Rollback Procedures

Always have a rollback plan:

1. •Know how to revert code deployment on your platform
2. •Know how to rollback database migrations
3. •Test rollback procedure in staging
4. •Document rollback commands

Post-Deployment Verification

After deployment:

1. •Verify health checks return 200 OK
2. •Test critical user workflows
3. •Monitor error rates and logs for 15+ minutes
4. •Verify database migrations applied correctly

Common Deployment Platforms

Choose platform based on your stack and requirements:

• •Static sites: Netlify, Vercel, CloudFlare Pages, GitHub Pages
• •Serverless: AWS Lambda, Vercel Functions, Netlify Functions, CloudFlare Workers
• •Containers: AWS ECS, Google Cloud Run, Azure Container Apps, Railway, Render
• •Virtual Machines: AWS EC2, DigitalOcean, Linode, Hetzner
• •Kubernetes: AWS EKS, Google GKE, Azure AKS, DigitalOcean Kubernetes

Consult your platform's documentation for specific deployment commands and configuration.

Rules

DO

• •Run full test suite before deploying
• •Validate environment configuration
• •Test production build locally
• •Document rollback procedure
• •Monitor after deployment
• •Tag releases in version control

DON'T

• •Deploy with failing tests
• •Commit secrets to version control
• •Deploy without rollback plan
• •Skip environment validation
• •Deploy breaking changes without coordination
• •Deploy on Friday afternoon without monitoring capacity

Integration

The orchestrator uses this skill when:

• •User requests deployment
• •Feature is complete and ready to ship
• •Rollback is needed
• •Environment setup required