Agent Skills: Deploy to AWS

Deploy to AWS

Deploy the Pitfal Solutions website following this sequence:

Pre-deployment Checks

1. Verify all tests pass: pnpm test
2. Verify TypeScript compiles: pnpm type-check
3. Check for uncommitted changes: git status

Build Process

1. Build the Next.js application: pnpm build
2. Verify the out/ directory was created

Infrastructure Deployment

1. Navigate to terraform: cd infrastructure/terraform
2. Initialize if needed: terraform init
3. Plan changes: terraform plan -out=tfplan
4. Show plan summary to user and confirm before applying
5. Apply changes: terraform apply tfplan

Static Site Deployment

1. Sync static assets to S3:

   aws s3 sync out/ s3://pitfal-static-site \
     --delete \
     --cache-control "public, max-age=31536000, immutable" \
     --exclude "*.html" \
     --profile pitfal
   

2. Sync HTML with no-cache:

   aws s3 sync out/ s3://pitfal-static-site \
     --cache-control "public, max-age=0, must-revalidate" \
     --include "*.html" \
     --profile pitfal
   

Post-deployment

1. Get CloudFront Distribution ID (if not already known):

   # Option 1: From Terraform output
   cd infrastructure/terraform && terraform output cloudfront_distribution_id
   
   # Option 2: From AWS CLI (finds distribution for pitfal.solutions)
   aws cloudfront list-distributions --profile pitfal \
     --query "DistributionList.Items[?contains(Aliases.Items, 'pitfal.solutions')].Id" \
     --output text
   
   # Option 3: Export as environment variable for reuse
   export DISTRIBUTION_ID=$(terraform -chdir=infrastructure/terraform output -raw cloudfront_distribution_id)
   

2. Invalidate CloudFront cache:

   aws cloudfront create-invalidation \
     --distribution-id $DISTRIBUTION_ID \
     --paths "/*" \
     --profile pitfal \
     --no-cli-pager
   

   Note: Cache invalidation takes 1-5 minutes to propagate globally.

3. Verify site is accessible at https://pitfal.solutions

4. Run smoke tests if available

Rollback

If deployment fails:

1. Check Terraform state: terraform state list
2. Restore previous S3 version if needed
3. Provide specific rollback instructions based on failure point