aws-deployment

name	aws-deployment
description	AWS deployment patterns and best practices for Amplify, App Runner, DynamoDB, S3, and Lambda. Use when deploying, configuring infrastructure, or designing cloud architecture.

AWS Deployment Patterns

Standard AWS deployment practices for all projects. Read the appropriate resource files for specific services.

Core Services Used

Service	Purpose	Resource
AWS Amplify	Frontend hosting (Next.js SSR)	resources/amplify-checklist.md
App Runner	Backend API hosting	resources/apprunner-patterns.md
DynamoDB	NoSQL database	resources/dynamodb-patterns.md
S3	File storage, static assets	resources/s3-patterns.md
Lambda	Serverless functions	See project-specific configs

Universal Rules

1. Least Privilege IAM: Every service gets its own IAM user/role with only the permissions it needs
2. Environment Variables: All secrets via AWS parameter store or environment config — never in code
3. Region Consistency: Default to us-east-1 unless project specifies otherwise
4. Tagging: All resources tagged with project, environment (dev/staging/prod)
5. Monitoring: CloudWatch alarms on error rates and latency

Instructions

1. Before ANY deployment, read the relevant resource file for the target service
2. Run the project's /preflight workflow to catch issues before deploy
3. Use the project's /courier workflow for the actual deploy process
4. Verify the deployment by checking health endpoints

Auto-Clone to Project

When you first detect a project uses AWS services:

1. Create .agent/skills/aws-deployment/ in the project root
2. Copy this SKILL.md as the base
3. Read the project's deployment configs (amplify.yml, apprunner.yaml, serverless.yml, etc.)
4. Append a ## Project-Specific section with discovered infrastructure details
5. Inform the user: "I've set up the local aws-deployment skill with your project's infrastructure patterns."