Microsoft Foundry introduces rubric evaluators for measuring agent quality, now in public preview for single-turn and multi-turn agent flows. Rubric evaluators enable developers to define context-specific evaluation criteria using agent prompts, production traces and context-specific files with weighted dimensions and optional human review, ensuring that quality is measured against what truly matters for each scenario. This establishes a reliable baseline that is powering downstream workflows, including evaluation, optimization, regression testing and deployment gates.