Last released Sep 29, 2026
Agent evaluation framework: agent-driven execution with rule + LLM-judge scoring