Agent Orchestration

Live pipeline view · OpenAI Agents SDK
Run complete in 47.3s
15 cases · 14 actions
Agent Pipeline — Sequential Execution
Agent 1
Schedule Monitor

Loads tomorrow's OR schedule. Identifies all cases requiring readiness review. Flags urgent and emergent cases for priority processing.

Agent 2
Readiness Reviewer

Checks labs, imaging, consent forms, and specialist clearances for each case. Scores readiness 0–100. Classifies as Ready, At Risk, or Blocked.

Agent 3
Care Coordinator

Creates targeted actions for every gap found: escalations to specialists, surgeon notifications, pre-op nursing alerts, and OR scheduling holds.

Agent 4
Briefing Generator

Produces concise clinical pre-operative briefings for each case including key risks, anesthesia considerations, and outstanding items.

Case Processing Results
Click to inspect per-agent traces
Agent Traces — Select a case
Step-by-step agent reasoning and tool calls
Select a case to view agent traces
Pipeline Performance Summary
47.3s
Total Run Time
15 cases end-to-end
3.2s
Avg Per Case
All 4 agents sequential
14
Actions Created
Dispatched automatically
6
Issues Caught
2 blocked · 4 at risk