Workflow evaluations gained clearer operating evidence

AI Infra evaluation paths now keep the intended tool configuration in view and retain a more complete timeline for review.

What changed

• Evaluation targets retain their configured tools and integration relationship during execution.
• Results record model activity, tool calls, ordering, and outcomes in one reviewable timeline.
• Operational views separate workflow, tool, credential, and persistence failures.

Why it matters

Teams can review one bounded workflow with evidence that reflects the configured execution path.

Scope: This improves evaluation evidence; it does not certify every workflow, tool, or integration for production exposure.