fhir
$
npx mdskill add mkurman/zorai/fhirQuery, create, and manage healthcare data via FHIR APIs.
- Enables agents to access patient records and clinical observations.
- Integrates with HAPI FHIR servers and major EHR platforms.
- Executes RESTful requests to retrieve or modify specific resources.
- Returns structured JSON responses containing parsed patient details.
SKILL.md
.github/skills/fhirView on GitHub ↗
---
name: fhir
description: "FHIR (Fast Healthcare Interoperability Resources) standard. Tools for reading, writing, and querying healthcare data via FHIR APIs: patients, observations, conditions, medications, procedures. Interop with EHR systems."
tags: [fhir, healthcare, ehr, interoperability, hl7, api, zorai]
---
## Overview
FHIR (Fast Healthcare Interoperability Resources) tools for reading, writing, and querying healthcare data via FHIR REST APIs. Work with Patient, Observation, Condition, MedicationRequest, and Encounter resources from EHR systems.
## Installation
```bash
uv pip install fhir.resources requests
```
## Query Patients
```python
import requests
from fhir.resources.patient import Patient
base_url = "https://hapi.fhir.org/baseR4"
resp = requests.get(f"{base_url}/Patient", params={"family": "Smith", "birthdate": "gt1970"})
data = resp.json()
for entry in data.get("entry", []):
patient = Patient.parse_obj(entry["resource"])
name = patient.name[0]
print(f"{name.family}, {name.given[0]}")
```
## Create a Resource
```python
patient = Patient(
name=[{"family": "Doe", "given": ["John"], "use": "official"}],
birthDate="1980-05-15",
gender="male",
)
resp = requests.post(
f"{base_url}/Patient",
json=patient.dict(),
headers={"Content-Type": "application/fhir+json"},
)
print(f"Created: {resp.json()['id']}")
```
## Workflow
1. Identify FHIR server endpoint (HAPI, Epic, Cerner)
2. Query resources with search parameters
3. Parse responses into typed resource objects
4. Create/update with POST/PUT
5. Use `$everything` for comprehensive patient data