September 13, 2026: Anthropic launched Claude smart reports in beta on September 10 for Claude Enterprise plans. The feature analyzes team activity to describe work produced, spending, obstacles and recurring patterns that could become shared skills, according to the official release notes.
What the beta includes
The smart reports guide says reports sample transcripts across chat, Claude Code and Cowork within a selected window of up to 28 days. Reports group work and costs, highlight friction, and suggest reusable workflows. During beta, organizations can generate 10 reports free per calendar month.
An Owner or Primary Owner must enable the feature, which is off by default. Organizations using customer-managed encryption keys, HIPAA configurations or Access Transparency are excluded, as are Claude Code for Enterprise organizations using zero data retention. Anthropic explicitly says reports should not be used for individual performance evaluation or employment decisions.
Why this matters for QA engineers
For QA leads, this creates a possible way to investigate where AI-assisted work stalls: repeated test-triage attempts, missing context or recurring setup problems. These are QATechTools’ suggested applications, not demonstrated improvements in defect detection.
A useful pilot would focus on one QA team and one reporting period. Compare the report’s themes with CI logs, reviewed bug reports and accepted test changes. A session that produced code does not establish that the code passed regression tests or met the requirement.
If repeated triage work emerges, turn one well-reviewed example into a shared workflow with required inputs and explicit validation steps. Measure whether it reduces rework on the next batch of failures before expanding its use.
What to check now
Enterprise QA leads should confirm eligibility and reporting access with their administrator, then choose a narrow pilot question such as which recurring setup issue consumes the most triage effort. Treat the beta’s findings as leads to investigate, and keep release decisions tied to independently verified testing evidence.