trace details

An observability capability in enterprise-level AI Agent platforms that can drill down to the specific nodes in an Agent's execution chain that cause slow responses, errors, or high costs, enabling administrators to adjust prompts, models, and tool choices in a targeted manner.