generated: '2026-08-14' method: generated source: >- openapi/cerner-millennium-fhir-r4-openapi.yml and openapi/cerner-millennium-bulk-data-openapi.yml — every operationId referenced by a skill was grepped out of those specs, which are themselves derived from the CapabilityStatement Oracle Health serves at /metadata. note: >- Four skills covering the flows that actually matter on this platform: read a patient, learn the contract with no credentials, pull a population as a backend service, and write back without creating duplicates. Each one is grounded in real operationIds and carries the Millennium-specific hazards — tenancy in the path, required search parameters, opaque paging that can repeat a resource, DataAbsentReason ambiguity, and the total absence of an idempotency primitive. skills: - file: cerner-patient-clinical-summary.md name: Build a patient clinical summary api: openapi/cerner-millennium-fhir-r4-openapi.yml operations: - searchPatient - readPatient - searchCondition - searchAllergyIntolerance - searchMedicationRequest - searchObservation - searchEncounter - searchProvenance - file: cerner-open-sandbox-exploration.md name: Explore Millennium with no credentials api: openapi/cerner-millennium-fhir-r4-openapi.yml operations: - getCapabilityStatement - searchPatient - readPatient - searchCondition - searchObservation - file: cerner-backend-bulk-export.md name: Run a bulk export as a backend service api: openapi/cerner-millennium-bulk-data-openapi.yml operations: - kickOffGroupExport - kickOffPatientListExport - getBulkExportJob - getBulkExportFileUrls - deleteBulkExportJob - file: cerner-write-back-clinical-resource.md name: Write back a clinical resource safely api: openapi/cerner-millennium-fhir-r4-openapi.yml operations: - createCondition - updateCondition - createAllergyIntolerance - updateAllergyIntolerance - createObservation - updateObservation - createDocumentReference - updateDocumentReference - patchPatient - readCondition