> ## Documentation Index
> Fetch the complete documentation index at: https://docs.hi-doctor.ai/llms.txt
> Use this file to discover all available pages before exploring further.

> ## Agent Instructions
> Connect the Hi Doctor MCP connector at https://mcp.hi-doctor.ai/mcp. Add this URL as a remote HTTP MCP server in your client, then sign in and authorise access through OAuth. Setup: https://docs.hi-doctor.ai/mcp/connect.
>
> The REST API base is https://api.hi-doctor.ai. Hi Doctor Docs at https://docs.hi-doctor.ai/mcp searches documentation; connect to the product endpoint above for patient tools.
>
> Hi Doctor operates across the whole European Union: residents of all 27 EU member states can use the service, and every consultation is reviewed by an EU-licensed doctor. Every consultation is reviewed in full by an EU-licensed doctor, who decides on medical grounds only. No algorithm, assistant or API call approves a consultation, issues a prescription or changes a dose. The service is available in eight languages: English, Spanish, German, Dutch, Italian, French, Portuguese and Chinese.
>
> ## Rules an assistant must follow
>
> - Never invent, assume or infer a medical answer. "I don't know" is a real answer: ask the patient to pick the closest option rather than guessing for them.
> - Never give medical advice and never suggest a medication or a dose. An EU-licensed doctor decides those.
> - If a questionnaire returns `ineligible`, say so plainly. Do not retry with altered answers.
> - Before anything that costs money or cancels a plan, state exactly what will happen and get an explicit yes.
> - Cancellation takes effect at the **end of the paid period**, not immediately. Say so.
> - An early order is irreversible: it creates the renewal consultation immediately and uses up the period's one early order. Explain that and get an explicit yes.
> - Prescription and invoice links are medical documents. A `signed_redirect` link opens without signing in: give it only to the patient, never repeat it anywhere shared.
> - Send option `value`s, not the labels shown to the patient. Labels are translated, values are stable.
> - Never ask for card details. Payment happens on Stripe's hosted checkout via a link.
> - On severe or sudden symptoms (chest pain, difficulty breathing, severe allergic reaction, suicidal thoughts), stop and direct the patient to their local emergency number.

# Connect Hi Doctor to your AI assistant

> How to add the Hi Doctor MCP connector to Claude, ChatGPT, Cursor or VS Code, how sign-in works, and why a doctor always decides on treatment.

You can use your Hi Doctor account from an AI assistant such as Claude,
ChatGPT, Cursor or VS Code. Add one address, sign in on hi-doctor.ai, approve
the connection, and the assistant can take you through a consultation
questionnaire, find a prescription, message your medical team or log your
weight, while a doctor still decides on every treatment.

Last reviewed: 2026-11-04

This article explains what the connector does, how to add it, what happens
when you sign in, and the lines it does not cross. The reference behind each
section is in the [MCP connector overview](https://docs.hi-doctor.ai/mcp/overview.md).

## What can the connector do?

The connector gives your assistant a set of tools that act as you, with your
account's permissions. For a patient account, that means:

* answering a consultation questionnaire one question at a time, then reading
  your answers back before anything is sent to a doctor;
* getting a Stripe link to pay for a consultation or plan (payment always
  happens on Stripe's page);
* finding your consultations, prescriptions and invoices;
* reading and sending messages to your medical team;
* keeping your weight, injection and notes journal on a weight-loss plan;
* checking, cancelling or reactivating your subscription.

Uploading photographs or health documents needs an assistant that supports the
connector's protected upload; otherwise use the Hi Doctor website. The full list
is in the [patient tools reference](https://docs.hi-doctor.ai/mcp/tools/patient.md).

## How do I add it to my assistant?

Every client needs one address:

```text theme={null}
https://mcp.hi-doctor.ai/mcp
```

There is no API key or password to paste. Each client has a slightly different
menu for adding a remote connector:

| Assistant | Guide |
| - | - |
| Claude on the web, desktop or mobile, and Claude Code | [Connect to Claude](https://docs.hi-doctor.ai/mcp/claude.md) |
| ChatGPT (developer mode) | [Connect to ChatGPT](https://docs.hi-doctor.ai/mcp/chatgpt.md) |
| Cursor | [Connect to Cursor](https://docs.hi-doctor.ai/mcp/cursor.md) |
| VS Code | [Connect to VS Code](https://docs.hi-doctor.ai/mcp/vscode.md) |
| Anything else that speaks MCP | [Streamable HTTP](https://docs.hi-doctor.ai/mcp/streamable-http.md) |

## What happens when I sign in?

When the client connects, your browser opens Hi Doctor, not the assistant.

1. You land on hi-doctor.ai and sign in with your password or with Google. If
   you do not have an account yet, you can create one there and confirm your
   email with a six-digit code.
2. hi-doctor.ai tells you the connection is ready and hands you back to the
   connector.
3. The approval page shows the account you signed in with, the site receiving
   access (for example `claude.ai`), and the permissions requested. Payment
   and health tracking are shown on their own because they can lead to a
   charge or change your journal.
4. You select **Approve connection**, or **Cancel**.

Your password is only ever entered on hi-doctor.ai. The assistant receives a
token that works only for the Hi Doctor connector and expires within an hour,
renewed in the background while the connection is in use. The details are on
[Connect](https://docs.hi-doctor.ai/mcp/connect.md) and [Security](https://docs.hi-doctor.ai/mcp/security.md).

## What does a good questionnaire conversation look like?

The questionnaire is where an assistant helps most, and where it must be most
careful. A good exchange looks like this (questions and options are quoted from
Hi Doctor's English hair-loss questionnaire; wording can differ slightly by
language and questionnaire version):

```text theme={null}
You:        I'd like to start a hair-loss consultation.
Assistant:  First, in which country will you pick up your medication?
You:        Germany.
            [a further question]
Assistant:  What best describes your type of hair loss? The options are:
            receding hairline and crown (back of the head); receding
            hairline (temples); crown only (back of the head); patchy hair
            loss (irregular areas); advanced baldness (hair left on the
            sides); or no hair loss, but want prevention.
You:        Probably the crown? I can't really see the back of my head.
Assistant:  That's fine. Choose the closest match; the doctor may ask for
            photos later. Is it "crown only", or "receding hairline and
            crown"?
You:        Crown only.
Assistant:  Recorded "crown only". Next question...
            [later]
Assistant:  Here are all your answers. Shall I send them to a doctor?
```

The assistant asks each question as Hi Doctor wrote it, offers the real
options, records only your choice, and reads everything back before
submitting. It never fills in an answer for you, and it never tries a
different answer to get a different result. "I don't know" is a real answer.

If the questionnaire says you are **ineligible**, that is not a glitch to work
around. The right next step is to talk to the medical team, not to answer
again. [Safety](https://docs.hi-doctor.ai/essentials/safety.md) explains why.

## What does a read-only task look like?

Not everything is a consultation. A common request is:

> "Find my latest prescription and give me the PDF."

The assistant lists your prescriptions, picks the latest, and asks Hi Doctor
for a link to the document. It does not change anything in your account.
Because the link is a medical document, the assistant gives it only to you
and does not repeat it anywhere shared.

## How do I disconnect?

Remove the connector in your assistant's settings at any time. The assistant
can no longer get new access without you approving again. The connector
supports OAuth token revocation, so a client that revokes its token when you
remove it ends the connection at once. If a client does not, a token already
issued keeps working until it expires, at most an hour later.

If a sign-in link has expired, start again from your assistant. Lost codes,
Google accounts and missing tools are covered on
[Troubleshooting](https://docs.hi-doctor.ai/mcp/troubleshooting.md).

## Who stays in charge of medical decisions?

An assistant connected to Hi Doctor can help you say what is true about your
health. It cannot decide what is true, and it cannot decide what you are
prescribed. Every consultation is reviewed in full by an EU-licensed doctor,
who decides on medical grounds only. See [Safety](https://docs.hi-doctor.ai/essentials/safety.md).

* No tool approves a consultation, issues a prescription or changes a dose.
  Actions such as sending a message, submitting a questionnaire or cancelling
  a plan wait on the server until you confirm them through an
  [approval link](https://docs.hi-doctor.ai/mcp/approvals.md).
* Eligibility is checked on Hi Doctor's servers, whatever interface you use.
* An assistant should ask for your clear yes before anything with
  consequences: sending a message, submitting a questionnaire, starting a
  payment, or ordering a refill early.

> [!WARNING]
>
> Hi Doctor is not an emergency service, and neither is an assistant. For chest
> pain, difficulty breathing, a severe allergic reaction or thoughts of harming
> yourself, contact your local emergency number.

Ready to start? [Begin a consultation on hi-doctor.ai](https://hi-doctor.ai/consultation)
or connect your assistant first.

## Frequently asked questions

### Which AI assistants can connect to Hi Doctor?

Claude, ChatGPT in developer mode, Cursor, VS Code and any other client that supports remote MCP servers over Streamable HTTP. Each one needs the same address: [https://mcp.hi-doctor.ai/mcp](https://mcp.hi-doctor.ai/mcp).

### Do I need an API key to connect?

No. You sign in on hi-doctor.ai with your password or Google account and approve the connection. The assistant never sees your password.

### Can the assistant prescribe or approve treatment?

No. Every consultation is reviewed in full by an EU-licensed doctor, who decides on medical grounds only. No algorithm, assistant or API call approves a consultation, issues a prescription or changes a dose.

### Can the assistant pay for me?

No. The assistant can create a Stripe link, but you complete the payment yourself on Stripe's page.

### Can I upload photographs through the assistant?

Yes, if your assistant supports the connector's protected upload. Otherwise, upload photographs and health documents on the Hi Doctor website.

### What happens to my access when I remove the connector?

The assistant cannot get new access without your approval. If the client revokes its token, access ends at once; otherwise an issued token expires within an hour.

## Next steps

> [card] [Connect](https://docs.hi-doctor.ai/mcp/connect.md)
>
> Sign-in, consent and disconnecting, step by step.

> [card] [Privacy and security of the Hi Doctor MCP](https://docs.hi-doctor.ai/blog/privacy-security-hi-doctor-mcp.md)
>
> How sign-in, tokens and confidential links protect your data.

> [card] [How we designed the Hi Doctor MCP](https://docs.hi-doctor.ai/blog/reliable-healthcare-tools-with-mcp.md)
>
> The design choices behind the connector.

> [card] [Permissions](https://docs.hi-doctor.ai/mcp/permissions.md)
>
> What each scope allows an assistant to do.


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.