Galileo has launched new agent-specific metrics aimed at enhancing user experience evaluations, expanding their Agent Evals MCP to include metrics accessible directly through an IDE. These new metrics—Agent Flow, Agent Efficiency, Conversation Quality, and Intent Change—complement an extensive suite of evaluation tools designed to improve AI infrastructure for clients like HP, Comcast, and NTT. The metrics assess how well agents adhere to workflows, execute tasks efficiently, maintain high-quality interactions, and handle user intent changes, which are crucial for optimizing user satisfaction and reducing infrastructure costs. Galileo's platform allows for custom domain-specific evaluations, offering a flexible and comprehensive agent evaluation framework to help AI teams build production-ready agents.