This project uses the AWS Strands Agent to orchestrate an AI-powered workflow for generating and deploying a Single Page Application (SPA) website. The agent automates the entire process:

1. Generate HTML
2. Generate Dockerfile & Nginx configuration
3. Build and push Docker image to AWS ECR
4. Generate Terraform configuration
5. Deploy the application to AWS ECS

Before you run this project, make sure you have installed and configured the following:

• Python 3.10.18
• uv Python package manager
• Docker
• Terraform
• AWS CLI configured with:
  • AWS Access Key ID
  • AWS Secret Access Key
  • Bedrock Model
  • IAM user permissions to:
    • Create and push ECR repositories
    • Provision ECS resources
    • Create VPC, Load Balancer, IAM Roles, and other infrastructure

1. Clone the repository

   git clone https://github.com/fromhell13/spa-web-builder-agent.git
   cd spa-web-builder-agent

2. Install dependencies

   uv sync

3. Activate Virtual Environment

   source .venv/bin/activate

4. Configure AWS credentials

   You can export them in your shell or configure them using aws configure:

   export AWS_ACCESS_KEY_ID=your-access-key
   export AWS_SECRET_ACCESS_KEY=your-secret-key
   export AWS_DEFAULT_REGION=your-region

5. Environment Configuration

   Rename .env.sample file to .env and fill up environment variable with your own data

   AWS_ACCESS_KEY_ID=
   AWS_SECRET_ACCESS_KEY=
   AWS_DEFAULT_REGION=
   AWS_ACCOUNT_ID=
   REPO_NAME=
   IMAGE_NAME=

6. Run the Agent Workflow

   uv run main.py

• Docker: Ensure Docker is running locally to build and push images.
• IAM Permissions: You need permissions for ECS, ECR, IAM, VPC, and related services.
• Region: Confirm your AWS region matches your ECR and ECS configuration.