AWS has expanded its QA Studio reference solution with application-level health metrics and team-based access controls. The practical change is that organisations can track test reliability across teams while limiting who can see or change each team’s applications.
AWS AI Watch analysis
What happened
The new capabilities organise tests and test suites under applications, each assigned to an owning team. QA Studio reports pass rates, execution totals and failures, and calculates a health score that penalises flaky tests. Teams can view trends over seven days at hourly granularity or 30 days at daily granularity, and filter metrics by environment.
Access is managed through five cumulative roles: Admin, Team Lead, Editor, Executor and Viewer. Teams can share applications with other teams at view, execute or full-access levels, but the recipient’s own role remains a ceiling on what they can do. The guide describes the capabilities as part of a reference solution deployed in the customer’s AWS account.
Why it matters
As AI-assisted testing spreads beyond one group, the awkward questions are less about writing another test and more about who owns it, whether it is healthy, and who is allowed to touch it. QA Studio’s metrics and application ownership give teams a way to answer those questions without treating a sprawling test list as a management strategy.
The access model also makes cross-team sharing more controlled: granting broad access to an application does not turn a Viewer into an Editor. AWS’s walkthrough gives concrete details of the design and its implementation with services including Amazon Cognito and CloudWatch.
Our read
This is a useful blueprint for teams trying to scale AI-powered QA without losing visibility or control. The strongest idea is the simple permission ceiling: sharing can widen access, but it cannot quietly outrank someone’s assigned role. It is a reference solution, not evidence that every organisation will get better test coverage by deploying it. Still, the health score’s flaky-test penalty is a sensible reminder that a green pass rate can flatter an unreliable suite.
What to watch
- Whether teams find the health score useful alongside the underlying pass, failure and flakiness figures.
- How application ownership and shared-access controls work in larger deployments.
- Whether AWS develops the reference solution further or customers report using it in production.
Discussion spark: Should AI-powered QA platforms prioritise cross-team sharing, or keep application access tightly siloed to preserve clear ownership?
Sources and evidence
- Accelerating software delivery with agentic QA automation using Amazon Nova Act – Part 3 (28 October 2026, 18:12 UTC)
not affiliated with or endorsed by Amazon Web Services (AWS)