AWS Introduces Preview of Well-Architected Agent

AWS recently announced the preview of the Well-Architected Agent, a service that analyzes environments against Well-Architected best practices and provides prioritized recommendations across cost, security, performance, and reliability. The AI agent can also suggest remediation steps and review infrastructure-as-code before deployment.

The AWS Well-Architected Agent correlates resource configurations, utilization metrics, application topology, and declared business goals to prioritize recommendations at the resource, application, and architecture levels, including trade-off analysis and remediation through CLI commands, IaC changes, and automation. The agent can also review Terraform and CloudFormation IaC before deployment. Channy Yun, lead blogger at AWS, writes:

It automatically correlates utilization metrics, resource configurations, and application topology, and analyzes against Well-Architected best practices across 65+ AWS services. It generates recommendations aligned to your declared business goals, delivers implementation packages with every finding, and surfaces cross-pillar trade-offs making it simpler to remediate the findings.

The Well-Architected Agent starts with an agent profile defining the accounts, regions, optimization pillars, and business goals. It then scans resources in the configured accounts and generates prioritized recommendations at the resource, application, and architecture levels for review and remediation. The agent analyzes infrastructure across cost optimization, security, resilience, and performance, and can also review IaC templates on demand.

The Well-Architected Agent uses application context and business goals to prioritize recommendations by impact and effort, providing findings at the resource, application, and architecture levels, including cost impacts and IaC changes. Yun adds:

You can also configure API access to integrate recommendations directly into your existing development and operations workflows. To interact with the agent programmatically, including calling APIs and searching documentation, try the AWS MCP Server and plugins with your preferred AI coding tool.

Some practitioners have questioned trusting an AI agent with architectural recommendations, while others point to practical challenges in prioritizing recommendations based on business goals. Nicolas Payette, founder and CEO of Specira AI, writes:

Genuinely curious where the agent reads those goals from, because in most enterprises nobody ever wrote down whether a given workload should trade cost for resilience, and that's exactly the call a recommendation has to make.

Alex Kargin, senior data engineer at Silicon Army, warns:

The part I’d watch is the jump from recommendations to automation-ready fixes. Cloud optimization is full of tradeoffs. Saving money can hurt resilience, and improving performance can increase cost.

While the agent represents an evolution of the popular AWS Well-Architected Tool for evaluating cloud architectures, AWS warns that its AI-generated recommendations may contain errors or incomplete information and should be evaluated in the specific context with appropriate oversight and safeguards. In his newsletter, Corey Quinn, chief cloud economist at The Duckbill Group, writes:

"Evaluates your environment as an experienced cloud architect would," is now being offered through the magic of AI® to accounts that can't get a solutions architect on the phone. If this thing works, what are 5,000 undifferentiated AWS "Partners" left to do?

The agent is currently available only in three US regions, although workloads from any region worldwide can be onboarded. Furthermore, it is available only to customers with a Business Support+ plan or higher.