Special offer. See more information about Outbyte and uninstall instructions. Please review EULA and Privacy policy.

iTechGuides is reader-supported. When you buy through links on our site, we may earn an affiliate commission. As an Amazon Associate I earn from qualifying purchases. Learn more

To add application context to AWS Well-Architected Agent, create a profile, configure its optimization goal, then attach an APPLICATION context with your workload’s scope and architecture. Check the profile’s eligibility and field errors before relying on scheduled recommendations. AWS currently labels the Agent as a preview feature, so confirm the latest documentation for availability and supported resources.

What application context does

AWS describes application context as additional workload information that helps the Agent generate more relevant, personalized recommendations. It supplements an agent profile; it does not replace one. The profile defines the monitored accounts and Regions, optimization goals and pillars, and application details. Context is added after profile creation.

The Agent offers three recommendation scopes: resource-level guidance for individual resources such as EC2, Lambda, or RDS; application-level guidance across related resources and their relationships; and architecture reviews of CloudFormation, Terraform, or AWS CDK templates before deployment, which can return updated templates. Choose the scope and pillars—cost optimization, security, resilience, and performance—that match the review you want.

Special offer. See more information about Outbyte and uninstall instructions. Please review EULA and Privacy policy.

AWS labels the Agent preview in its concepts and terminology documentation. Supported features and resource types may change; consult that documentation before assuming a resource is covered.

What to include in application context

Describe the actual workload and the discovery boundary, rather than using broad instructions such as “optimize everything.” AWS does not publish a required wording template or promise that a particular phrase will change a recommendation.

  • Application overview: Say what the application does and identify the workload the profile should analyze.
  • Accounts and Regions: Provide the target AWS account IDs and Regions. Use resource tags to narrow discovery where appropriate.
  • Services and resource types: Identify relevant AWS services and resource types. Check current supported-resource documentation because some types are unsupported.
  • Application classification: Supply the industry, application type, and criticality that accurately describe the workload.
  • Architecture overview: Explain the major components and how they connect.
  • Additional context: Add material constraints or optimization priorities that the defined fields do not capture.

The API field names are applicationOverview, accountIds, regions, resourceTags, awsServices, resourceTypes, industry, applicationType, criticality, architectureOverview, and additionalContext. The AWS CLI reference allows up to 10,000 characters for application overview, architecture overview, and additional context; see the create-agent-context reference for the current field definitions and constraints.

Set up the profile and add context

Follow AWS’s documented sequence: configure the roles, create the profile with its account configuration and target pillars, add an optimization goal, create application context for the profile, then check eligibility. The API quickstart provides the full setup workflow.

Special offer. See more information about Outbyte and uninstall instructions. Please review EULA and Privacy policy.
  1. Configure execution and access roles. Set up the execution role in the profile account and access roles in the target accounts, as described below.
  2. Create the agent profile. Choose the target accounts, Regions, and pillars to review.
  3. Add an optimization goal. Configure at least the goal required by your workflow.
  4. Create application context. Associate it with the profile using the APPLICATION context type, a title, and typed content. Replace the sample values in this AWS CLI command with your workload’s actual details:
aws wellarchitected create-agent-context 
  --profile-arn "arn:aws:wellarchitected:us-east-1:111122223333:agent-profile/my-production-profile" 
  --title "Payment processing service" 
  --context-type APPLICATION 
  --content '{"applicationOverview":"Real-time payment processing","criticality":"MISSION_CRITICAL","accountIds":["444455556666"],"regions":["us-east-1"]}'

The overview, mission-critical value, account ID, and Region shown are illustrative values in AWS’s example, not defaults.

  1. Verify profile eligibility. Run get-agent-profile for the profile and inspect eligibleForScheduledGeneration and fieldErrors. If eligibility is false, use the reported field errors and AWS troubleshooting guidance to correct the profile or context. When eligibility is true, AWS says scheduled recommendations begin within 48 hours; this is the documented workflow timing, not a guarantee of recommendation quality or a measured performance result.
Independent reader supportYour contribution helps us test, update, and keep practical guides available for everyone.Support on Ko-Fi

Check permissions and ongoing access

The Agent uses customer-managed IAM roles. An execution role in the profile account orchestrates discovery by assuming access roles in target accounts. AWS says those access roles provide read-only access to resource metadata and configuration. Its recommended managed scanning policy can be replaced with a custom policy scoped to the actions required. Review the AWS Well-Architected Agent access model before deployment.

  • Consider running profiles from a dedicated AWS account without production workloads.
  • Use the execution role ARN in access-role trust policies.
  • Monitor AWS CloudTrail for Agent API activity.
  • Review permissions periodically and remove accounts that no longer need analysis.
  • Retain the ability to revoke access by removing trust, deleting an access role, or revoking active sessions.

What to verify before acting on recommendations

  • Confirm that the profile’s accounts, Regions, tags, services, and resource types match the intended discovery scope.
  • Check whether each resource type is currently supported rather than assuming the Agent covers every workload component.
  • Verify eligibleForScheduledGeneration and resolve any fieldErrors before waiting for scheduled results.
  • Choose resource, application, or architecture scope and the relevant pillars for the decision at hand.
  • Review the recommendation against your workload’s architecture and constraints; context is intended to improve relevance, not to guarantee a specific result.

Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.