The current built-in Agent Analytics are very surface level and the insights they provide are really about ROI. I want insights that provide me when my critical ai workflows fail, why X run used crazy amount of credits, what exact tools calls were used etc. It would be great if we can export super agent run details via API or even if you supports an otel export. Either way I find this to be a blocker from migrating from other AI orchestration tools fully specially for critical workflows.