| Resource | Description |
|---|---|
| Environments | Simulated health systems composed of linked simulators |
| Simulators | Reusable simulator definitions (FHIR, HL7, Voice, Fax, etc.) |
| Datasets | Synthetic patient data (FHIR bundles, files) |
| Benchmarks | Groups of tasks with evaluation criteria |
| Tasks | Individual test cases within benchmarks |
| Evals | Natural language assertions on tasks |
| Playgrounds | Running environment instances with live sandboxes |
| Sandboxes | Running simulator instances with credentials |
| Runs | Benchmark executions with scores and verdicts |
| Task Runs | Task-level results within a run |
| Eval Runs | Eval-level results within a task run |