aws-samples/well-architected-iac-analyzer

What it solves

This project provides a way to automatically evaluate Infrastructure as Code (IaC) templates and architecture diagrams against the AWS Well-Architected Framework. It helps users identify where their cloud infrastructure deviates from best practices and provides actionable recommendations for improvement.

How it works

The application uses a React frontend and Amazon Bedrock to analyze uploaded files—such as CloudFormation, Terraform, AWS CDK templates, or architecture diagrams (PNG/JPG)—against a knowledge base of AWS Well-Architected whitepapers. It employs a RAG (Retrieval-Augmented Generation) approach, utilizing S3 Vectors for cost-effective storage of best practices. The system can process multiple files in parallel to speed up analysis and supports various LLMs (specifically the Claude family) to reason through complex architectural designs.

Who it’s for

Cloud architects, DevOps engineers, and security professionals who want to ensure their AWS infrastructure is compliant with industry best practices and organizational standards.

Highlights

• Multi-Format Support: Analyzes CloudFormation, Terraform, CDK, PDF documentation, and image-based architecture diagrams.
• AI-Powered Guidance: Includes an interactive chatbot assistant for personalized guidance and clarification on analysis results.
• Prioritization Framework: Uses an Eisenhower Matrix to plot "Not Applied" best practices by risk criticality and implementation effort.
• Flexible Analysis: Supports official AWS lenses (e.g., Serverless, IoT, Machine Learning) and allows for the creation of custom organizational lenses.
• Template Generation: Can generate IaC templates directly from uploaded architecture diagrams.