fhir-synth

Generate synthetic FHIR-compliant healthcare datasets from natural language prompts.

Updated Feb 9, 2026
One-click install
npx skills add https://github.com/alvinhenrick/fhir-synth --skill fhir-synth
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: fhir-synth
Source: https://github.com/alvinhenrick/fhir-synth/tree/main/src/fhir_synth/skills/builtin/coverage
Command: npx skills add https://github.com/alvinhenrick/fhir-synth --skill fhir-synth

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill addresses the need for creating realistic synthetic healthcare data efficiently, enabling testing, analysis, and development without compromising patient privacy.

Core Features & Use Cases

  • Data Generation from Prompts: Converts natural language descriptions into synthetic FHIR-compliant datasets, including patients, conditions, and observations.
  • Programmatic Workflow: Supports complex scenarios such as batch generation, version selection (R4B or STU3), and adding follow-up encounters.
  • Use Case: A healthcare developer needs to simulate a patient population with specific attributes to test an EMR system or run analysis, and can generate such data instantly by describing the scenario in plain English.

Quick Start

Describe your synthetic data needs, such as "generate 10 diabetic patients," and the system will produce the data accordingly.

Frequently Asked Questions about fhir-synth

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I generate synthetic FHIR healthcare data from a natural language prompt?

Synthetic FHIR data generation works by converting natural language descriptions into programmatic code that produces FHIR-compliant resources like patients, conditions, and observations without using real patient data.

Can I generate synthetic clinical data for both FHIR STU3 and R4B versions?

Yes, you can generate synthetic clinical data for both FHIR STU3 and R4B versions. The system supports programmatic version selection to ensure the generated resources match your specific EMR or testing environment requirements.

Do I need Python and FHIR resource libraries to create synthetic healthcare datasets?

Yes, creating synthetic healthcare datasets requires Python, FHIR resource libraries, and LLM integration for code generation to automate the production of realistic, privacy-preserving clinical records.

What's the best way to add follow-up encounters to an existing synthetic patient population?

The best way to add follow-up encounters is by using the programmatic workflow feature, which handles complex scenarios like batch generation and appending follow-up data to an existing synthetic patient population.

Why use synthetic healthcare data instead of real clinical records for EMR testing?

You use synthetic healthcare data instead of real clinical records to enable EMR testing and analysis without compromising patient privacy, ensuring compliance while still simulating realistic patient attributes and conditions.