Build a FHIR Care-Team On-Call Task Routing Engine — implementation

## Build
Route urgent research protocol tasks to on-duty study coordinators using FHIR R5 Task and CareTeam structures. Audit assignment latencies and log message transfers across institutional shifts.

## User
Clinical Trials Operations Specialist

## Workflow
1. Ingest protocol milestone alerts formatted as FHIR R5 Task objects.
2. Retrieve current on-call personnel shifts from a CareTeam schedule registry.
3. Apply availability, role qualifications, and shift boundaries to route task ownership.
4. Generate updated FHIR Task resources with assigned performers and delivery confirmation timestamps.

## Interface
Task assignment dashboard with live shift timelines, open protocol tickets, and re-routing triggers.

## Data
Synthesize FHIR R5 CareTeam, Practitioner, and Task resources following HL7 FHIR Foundation specifications.
Starter examples below are synthetic. They are not downloaded records.

## Source
[HL7 FHIR overview](https://hl7.org/fhir/overview.html)
Reference FHIR R5 Foundation CareTeam participant types and Task status codes to maintain standard conformance.

## Tools
semantic-kernel, pydantic, fastapi.
semantic-kernel orchestrates deterministic skills for shift schedule parsing and task assignment execution.
Hosted or local model connector supported by Semantic Kernel to synthesize handover notes.

## Deliverables
Submit runnable source, pinned dependencies, setup commands, a working interface, example outputs, and a recorded end-to-end demo. Include tests and an action log.

## Acceptance
- Consumes valid FHIR Task objects and parses current status and priority codes.
- Matches task domain to eligible active practitioners within the active shift schedule.
- Produces updated Task payload containing valid reference to assigned Practitioner.
- Maintains an immutable routing decision log for operational compliance audits.

## Starter Examples
```json
[
  {
    "input": {
      "task_id": "TASK-BIO-01",
      "priority": "urgent",
      "focus": "SpecimenCollectionReview",
      "care_team": {
        "id": "TEAM-ONC",
        "members": [
          {
            "practitioner_id": "PRAC-A",
            "role": "coordinator",
            "status": "on_shift"
          },
          {
            "practitioner_id": "PRAC-B",
            "role": "coordinator",
            "status": "off_shift"
          }
        ]
      }
    },
    "expected_output": {
      "task_id": "TASK-BIO-01",
      "status": "assigned",
      "assigned_to": "PRAC-A",
      "escalated": false,
      "routing_reason": "practitioner_active_on_shift"
    }
  },
  {
    "input": {
      "task_id": "TASK-BIO-02",
      "priority": "stat",
      "focus": "AdverseEventAudit",
      "care_team": {
        "id": "TEAM-ONC",
        "members": [
          {
            "practitioner_id": "PRAC-B",
            "role": "coordinator",
            "status": "off_shift"
          }
        ]
      }
    },
    "expected_output": {
      "task_id": "TASK-BIO-02",
      "status": "held",
      "assigned_to": null,
      "escalated": true,
      "routing_reason": "no_active_member_on_shift"
    }
  }
]
```

## Safety
Research administration operational simulator; operates on synthetic practitioner data; no emergency medical triage.
Use mock external writes. Require human review before any real deployment.