Resources · Claude connector

Claude Function Health connectorWhat Claude can do in your Function Health account.

The Claude Function Health connector exposes 3 tools. 3 read, 0 write, 0 left undocumented. Claude sees a high-level view of your lab results and the nutrition plan you built in Function, never the detailed values. Here is what it reads, what stays out and who can revoke it.

Verified Trustpilot reviews · AI, automation & growth agency

Overview

What changes when you read your results

Your Function results already sit in the Function app. The connector lets you talk them through with Claude: which health areas show out-of-range markers, how the categories compare, and what your nutrition protocol says, all in the same conversation where you can ask follow-up questions in your own words.

Get the big picture in one question. Ask how your results look overall and Health Summary gives Claude the total number of biomarkers, how many are in range and how many are not.

Zoom in on one area. Worried about heart health after a family history chat? Biomarkers returns the in-range and out-of-range counts for the Heart category, or for every category if you ask for the full picture.

Turn your plan into a shopping list. Health Action Plan / Meal Plan brings back the nutrition protocol you created in Function, with its goals and meal ideas, so Claude can help you plan the week's groceries around it.

What it will not do: Function says the connector is read-only and cannot change anything in your account. It also never passes individual test values, biomarker identifiers, your email address or detailed personal health information. So Claude can tell you that two heart markers are out of range, not which ones or by how much. And nothing runs by itself: no alert fires when new results land. Other apps and platforms are listed on our Integrations page; if you log habits in a spreadsheet, the Claude airtable connector can sit alongside this one in the same chat.

Vocabulary

The vocabulary in one minute

Five words you will see while connecting Function Health.

Connector
The link you set up once between Claude and an account you already have, so Claude can use it while it answers you.
Tool
One named thing a connector lets Claude do. Claude picks the tools it needs by itself, and the directory lists them by name.
Authorization
The service's own sign-in and consent screen, where you grant the access Claude will use. Given once per person and revocable.
Approval
The confirmation Claude waits for before it goes ahead with something that changes an account, shown in the chat when it matters.
MCP
The shared standard connectors are built on: it is what lets an assistant like Claude talk to an outside service like Function.
Connect

Plug Function Health into Claude in three steps

  1. 01

    Find Function Health in Claude

    In Claude's settings, open Customize, then Connectors, and find Function Health in the list. On a Team or Enterprise workspace, an Owner or Primary Owner must turn the connector on before members can sign in with their own account.

  2. 02

    Start the connection

    Click Connect on its row, then log in to Function in the window that Function opens itself. If the link ever breaks, use Disconnect, then connect again.

  3. 03

    Review the permissions

    Read Function's authorization screen before you accept. It belongs to Function, not to Claude, and it sets what the access covers. You can also withdraw that access later from your Function account.

Tools

The 3 tools, sorted by what they do

Function Health gives Claude 3 tools: 3 that read your account, 0 that change something in it.

All three read summary data from your Function account. Their names stay exactly as Claude displays them.

  • 3 read

What Claude reads (3)

3 tools

Three read-only views of your results and your nutrition protocol.

Health Summary

Gives Claude a bird's-eye count of your lab results: the total number of biomarkers and how many sit in range versus out of range.

When it helps
your results just came in and you want the headline before reading anything else.
Watch out
you get counts, never the individual values behind them.

Sourceservices.functionhealth.com · September 30, 2026 ↗

Biomarkers

Breaks the results down by health area, such as heart, blood or kidney, with in-range and out-of-range counts for each area.

When it helps
you want to know which areas deserve a question at your next doctor's appointment.
Watch out
Function lists completed lab results as a prerequisite, so an account without them gives Claude nothing to break down.

Sourceservices.functionhealth.com · September 30, 2026 ↗

Health Action Plan / Meal Plan

Retrieves the nutrition plan you set up in Function, with your goals and meal ideas, so Claude can reason from it.

When it helps
it is Sunday evening and you want a grocery list that fits the plan.
Watch out
Function grants read access only here, so Claude can build around the plan but cannot edit it in your account.

Sourceservices.functionhealth.com · September 30, 2026 ↗

Approvals

What Claude asks before acting

By default, Claude stops and asks for your go-ahead before each action it takes on an account for you. Function describes its connector as read-only, so none of its three tools changes anything for that rule to catch.

On Team and Enterprise workspaces, owners decide whether members can let some actions through without being asked every time, and they can restrict what a connector may do for everyone. Claude works with the access of the person who connected and nothing more: what that person cannot open in Function stays closed to Claude too. The real gate here is Function's consent screen, where you grant three read permissions one by one.

Plans

Which plans include it

Of the 819 sheets in the official directory, none displays availability by plan. The answer connector by connector is published nowhere, which is a real gap in the catalogue.

The general rule is published: remote connectors are open to all users on Claude, Cowork, Claude Desktop and mobile. On Team and Enterprise, an Owner or Primary Owner turns the connector on for the organisation first. Function's own guide says something narrower, shown side by side in the limits section. The connector's sheet in the official directory is where its current state appears.

Limits

Where this connector stops

A connector is not an automation. Claude calls these tools while it answers you: nothing is sent to you when new results arrive in Function.

By default, the sheet only exposes reading. That is an observed floor rather than a guarantee, since an admin can open actions no public sheet shows. The partner badge is not a security audit, and Anthropic says on every sheet that it does not decide which tools a publisher exposes. Everything here comes from Function's documentation: neither Claude's help centre nor Anthropic's developer docs cover this connector.

Two official sources disagree

Which Claude plans can use the Function Health connector?

What this page followsNeither page is marked out of date, and they answer different questions: Anthropic's rule covers every connector, Function's line targets this one but comes from the publisher, on an undated page. Neither is final for your account: open the Function Health sheet in the directory while signed in to Claude and check there.

Need help

Need help connecting Function Health to Claude?

A person reads every message.

FAQ

FAQ about the Claude Function Health connector

01What can Claude do with the Function Health connector?
Claude can read a high-level summary of your Function lab results and the nutrition plan you created in Function. The connector exposes three tools: an overall count of in-range and out-of-range biomarkers, a breakdown of those counts by health category such as heart or kidney, and your nutrition protocol with goals and meal ideas. You can then ask Claude which areas deserve attention or build a grocery list around your plan.
02Can Claude change anything in my Function account?
No. Function describes the connector as read-only and says it cannot modify your information. All three tools only read summaries: overall counts, counts by category and your existing nutrition protocol. Claude can draft ideas from them in the chat, but nothing it writes goes back into Function. As on any connector, the list of tools on a public sheet is an observed floor, but Function's own guide is explicit about read-only access here.
03Does Claude ask before it reads my health data?
No source describes a confirmation specific to the Function tools. The general rule is that Claude asks before each action it takes on an account for you, and here every tool only reads. The decisive moment is earlier: Function's consent screen, where you approve three read permissions covering your results summary, your category counts and your nutrition plans. You can revoke that authorization at any time from Claude's settings or from your Function account.
04Which Claude plans is it available on?
No official source publishes plan availability connector by connector, and no directory sheet displays it. Anthropic's general rule is that remote connectors are open to all users on Claude, Cowork, Claude Desktop and mobile, with an Owner or Primary Owner turning them on first on Team and Enterprise. Function's guide states a narrower prerequisite, quoted in the limits section. The Function Health sheet in the official directory is the only place showing the current state for your account.
05Does Claude see my individual lab values?
No. Function states that the connector retrieves summary-level data only: test counts, category names with aggregate figures, nutrition goals and meal plans. Individual lab values, biomarker identifiers, email addresses and detailed personal health information are never sent. Claude can therefore tell you how many markers are out of range in a category, but not which marker or what the number was. For those details, the Function app remains the place to look.
06Why does Claude say it found no results?
A first thing to check is whether your Function account already holds completed lab results: Function's guide lists them as a prerequisite, so without them the three tools have nothing to summarise. The other case Function documents is an "Insufficient Permissions" error, when the required scopes were not granted at sign-in. Its fix is to disconnect the connector in Claude, connect it again and approve every permission requested on the new authorization screen.
07Claude or an automation tool for Function Health?
They serve different uses. Claude with this connector is for a conversation about your results: you ask, it reads the summary and explains, you follow up. It never runs on its own, so it cannot notify you when new results arrive or copy figures somewhere on a schedule. Work that has to happen by itself in the background belongs to an automation platform, a different kind of tool rather than a better or worse one.