For developers

Free FHIR server for development: a sandbox with 30 synthetic patients and six months of history

Where to get a free FHIR R4 server you can write to today: the Anpheros sandbox, self-service, with 30 synthetic patients, realistic LOINC, ICD-10 and ATC data, a REST API, webhooks and SDKs. Limits, resets and what the data looks like.

by Published 3 min read

Public FHIR test servers are useful for a first request and useless the moment you need data that stays put, patients that look like yours, or a write that does not get wiped by the next person. The Anpheros sandbox is a free FHIR R4 server that belongs to your project: your own 30 synthetic patients, your own writes, your own keys and webhooks, and a reset button when you want the original data back.

What you get

  • A FHIR R4 endpoint at https://platform.anpheros.com/fhir/R4, with a CapabilityStatement at /metadata, search, _history, $everything, $summary (International Patient Summary) and $validate.
  • The same data over REST at /v1: patients, observations, medications, conditions, timeline, documents.
  • 30 synthetic patients per project, with six months of history up to the day the project was created: adults, children and older people from across the EU, with conditions coded in ICD-10, medications in ATC, allergies, lab results and vital signs in LOINC.
  • Webhooks, documents, lab import, OAuth, exactly as in production, so nothing changes when you go live.
  • SDKs for Dart (anpheros_sdk on pub.dev) and TypeScript (@anpheros/sdk on npm).

The sandbox is free and stays free. Nobody asks for a card.

Getting a key

Sign in to the dashboard at platform.anpheros.com/dashboard with Google and press Get a sandbox key. You receive an organization, a sandbox project and a sk_test_… key, shown once. The same through the Admin API:

curl -X POST https://platform.anpheros.com/admin/sandbox/quickstart \
     -H "Authorization: Bearer $FIREBASE_ID_TOKEN"

First call:

export BASE=https://platform.anpheros.com
export KEY=sk_test_…
curl "$BASE/fhir/R4/Patient?_count=5" -H "Authorization: Bearer $KEY"
curl "$BASE/fhir/R4/Observation?patient=$PID&category=laboratory&_sort=-date" -H "Authorization: Bearer $KEY"

The synthetic patients

Each sandbox project gets its own copy. Change or delete anything; no other project sees it. Reset test patients in the dashboard (or POST /admin/projects/{id}/sandbox/reset) brings them back as they were, with new ids; patients you created yourself are not touched. Each synthetic patient keeps a stable identifier across resets, so a test can always find the same person:

Patient?identifier=https://anpheros.com/fhir/sid/sandbox-persona|p02

is always Elena Ionescu, born 1985, with the same conditions and the same six months of readings.

Limits

Generous but finite, so one runaway script cannot cost anyone anything:

Limit Sandbox
Resource writes 10 000 per project per day
Patients 500 of your own, besides the 30 synthetic ones
Documents 250 MB per project
Requests 600 per minute per key
Sandbox projects 5 per organisation

A sandbox key only ever sees sandbox data. Production is a different database, opened to verified organisations with a signed data processing agreement; the first month of production is free.

Browsing the data

The dashboard shows every patient's record, the FHIR of each resource, the Provenance behind it and the project's request log. For a quick look without code, GET /v1/patients/{id}/timeline returns everything about one person in date order, and GET /fhir/R4/Patient/{id}/$everything returns the whole record as a Bundle. Example resources for every type are published at platform.anpheros.com/tools/fhir-examples, and a validator at /tools/fhir-validator.

Frequently asked questions

Is the sandbox really free?

Yes, with no time limit and no card. The limits above are there to keep it free for everyone.

Is it a public FHIR server?

No, and that is the advantage. Public test servers are shared and wiped; your sandbox belongs to your project, with its own keys. Other developers cannot see or change your data.

Can I load my own test patients?

Yes: create patients and resources through /fhir/R4 or /v1, up to 500 patients per project. The lab connector imports a whole report from CSV, HL7 v2 or JSON in one request.

What happens when I go to production?

Nothing changes in your code. The base URL is the same; a sk_live_… key points the same calls at the production database. Production requires a verified organisation and a signed data processing agreement.

Can an AI assistant use the sandbox?

Yes. The MCP server at platform.anpheros.com/mcp accepts a sandbox key and exposes the record to any MCP-capable assistant: list patients, get a provenance-labelled context, search FHIR, validate and create resources. See EMR integration with AI medical assistants.

This guide is for information and does not replace medical advice. For decisions about your health, talk to your doctor. In an emergency, call 112.
About the author
Adrian Kereky

Founder of Anpheros. An electronics engineer, he spent six years in autonomous driving, on automotive projects run in partnership with Samsung and NVIDIA, designing the architecture of driver-assistance and AI systems that run on NVIDIA chips with CUDA. He then built Anpheros: the patient-controlled medical record and the HL7 FHIR R4 platform it runs on.

More about Anpheros and the author →

Keep your medical record with you

Anpheros Daily keeps symptoms, lab results, medications and documents in one place, in a standard format, under your control. The Basic plan is free.

Get Anpheros Daily