About

About this project

Cloud Agent Lab is a CSCE 412 (Cloud Computing) project at Texas A&M University, Fall 2026. The task: engineer a multi-agent AI system that helps plan, deploy, secure and verify a real website on AWS.

Goals

Host a multi-page static site on a custom domain over HTTPS
Keep the storage bucket private behind CloudFront
Make deployment repeatable with scripts, not console clicks
Detect, diagnose and recover from a real configuration failure

Tools

Python with the AWS SDK (boto3)
AWS CLI with a limited-permission IAM user
An OpenAI-compatible LLM API for agent reasoning
Git and GitHub for source control
Namecheap for domain registration

AI assistance

Following the course's AI policy, AI coding assistants were used to help write the deployment scripts, the agent code and this website. All code was reviewed, run and tested against the live infrastructure.

Frequently asked questions

Why is there no web server?

The site is entirely static: HTML, CSS, JavaScript and images. CloudFront serves those files directly from S3, so there is nothing to patch, scale or keep running.

Why not turn on S3 static website hosting?

The S3 website endpoint only works with a public bucket and does not support HTTPS. Using CloudFront with origin access control keeps the bucket private and the connection encrypted.

Can the agents change the infrastructure on their own?

No. Agents can read configuration and run checks, but changes that could break the live site wait for a person to approve them.

What does it cost to run?

Very little: the Route 53 hosted zone costs $0.50 per month, the ACM certificate is free, and S3 and CloudFront usage at this traffic level costs a few cents at most.