1health developer registration

One registration. A running healthcare app to learn from.

Register without a credit card, make your first 1health API call in seconds, and preview your own secure Patient Vault — a live demo application built on the platform, with source code and per-endpoint agent docs.

Free · Demo environment · You'll land in Patient Vault

What happens when you register

Three steps, no sales call

The button above takes you directly to the registration screen. Here's exactly what you're signing up for.

01

Register — no credit card.

One registration creates your 1health developer account: an organization and user in the demo environment, the same platform infrastructure our team ships enterprise apps on.

02

Make your first API call in seconds.

You land in Patient Vault — a demo application. Click through its UI and the API Inspector shows the real 1health calls as you make them: request, response, live against the platform.

03

Preview your own secure Patient Vault.

Your registration provisions a vault of your own — store a patient, attach a document, find them again. Then read the source on GitHub to see exactly how the app does it.

Patient Vault

DEMO APPLIVE

The patient database for your healthcare app — running in production as a working example of a 1health platform app.

Three verbs. One mental model.

Every endpoint and every shape fits into one of three verbs.

STORE

Create and update patient records: demographics, identity, the canonical patient.

POST /patient

ATTACH

Add anything to a patient: addresses, contacts, documents, observations, wearable data, arbitrary attachments.

POST /patient/{id}/addressPOST /patient/{id}/contactPOST /patient/{id}/identifierPOST /patient/{id}/document

FIND

Locate patients in your vault. One endpoint, one boolean: exact true for a definitive lookup, false for probabilistic matching.

GET /patient/findGET /patient/{id}/document

Reading is just GET — the full record, paginated, or a metadata-only manifest.

Log in to Patient VaultLaunch the demo app on pv.demo.1health.ioView source on GitHubTachin-ai-Corporation/patient-vault-officialAPI docs — every endpoint & verbagents.md · examples.md per route
API Inspector — fires as you click the UI
POST /patient                 201 Created
POST /patient/{id}/document   201 Created
GET  /patient/find?exact=true 200 OK

Then make it yours

From demo to your own app

Don't start from a blank page. Bootstrap from a repo that's already wired to the platform.

  1. 1

    Pull the bootstrap repo.

    Clone patient-vault-official from GitHub — auth, API client, and platform conventions already in place.

    Open repository →
  2. 2

    Drop it into Vercel.

    Import the repo, add your keys from registration, deploy. You have a running 1health-connected app in minutes.

  3. 3

    Point your agent at it.

    Use a prompt like the one here — your coding agent reads the per-endpoint docs and builds against real platform APIs.

Example prompt — v0 / Claude / Cursor
I'm building a healthcare app on the 1health platform. Start from my
imported 1health bootstrap app.

Read the agents.md file for each API endpoint I use before calling it.
Build a simple intake screen that registers a patient, attaches a
referral PDF, and lists that patient's documents.

Use POST /patient, POST /patient/{id}/document and
GET /patient/{id}/document. Use the simplest APIs that accomplish
the task.

Same flow works on Replit, Lovable, and GCP.

Documentation built for agents

Two files per endpoint. Two ways to consume them.

Every 1health API endpoint and verb is documented in markdown that coding agents read natively — regenerated automatically as the platform evolves.

agents.md

One file per route: what each endpoint and verb does — overview, authorization, parameters, request body, responses, curl. Generated from platform source, so the docs can't drift from the API.

examples.md

A sibling file per route: real usage patterns extracted from production apps built on 1health — verbatim code, the resolved request, response handling, and the quirks that aren't in the reference docs.

No MCP server required

Path A — Read the files directly

/api/v3/patient/
├── agents.md · examples.md
├── _patientId_/
│   ├── address/ agents.md · examples.md
│   ├── attach/  agents.md · examples.md
│   └── contact/ agents.md · examples.md
└── find/ agents.md · examples.md

A browsable tree of markdown, one node per route — point any agent at a URL, or pull the whole tree locally. Start at the manifest for one line per endpoint, or browse by task. Coverage today spans every endpoint observed across five production applications, and is expanding.

Best for conversational building

Path B — Connect the MCP server

One endpoint your agent connects to. Ask ‘I want to build X — how do I do it?’ and the server searches the same knowledge base, finds the closest patterns from real apps, and answers with the specific APIs to use. Depending on your agent, model, and effort settings, one path will work better than the other — try both.

https://mcp.dev.1hdev.io/mcp

Questions before you start

Land in a working app two minutes from now

Register, explore Patient Vault, read its source, then ship your own.

Create your developer account →