Health Data (your own Supabase)

io.github.almostjackedPublished by GitHub account almostjackedv0.1.3Updated Oct 24, 2026

Your health data (weight, macros, weigh-ins) as MCP tools — in your own Supabase, zero custody.

The publisher's own server: Supabase

VerifiedSTDIODesktop onlyDatabasesLocation & LifestyleData & Analytics

Overview

AI-generated overview

A local MCP server that lets an assistant read your own Apple Health data (weight, calories, macros, water) stored in a Supabase project you create and own.

What you can do
Inspect the database schema and metric registry before querying, and check sync status to see whether the daily import has stalled. Read recent rows over a chosen number of days, optionally for one metric, and get a full per-day summary of totals and measurements. Compute min, max, average and rolling-average series for one metric over a date range with daily, weekly or monthly rollups. Run one read-only SQL statement as an escape hatch, with writes rejected and a row limit enforced. Export daily date, weight and calorie entries shaped for pairing with a separate TDEE tool.
When to use it
Worth adding if you already track weight, calories or body composition in Apple Health and want an assistant to answer trend and intake questions from your own database. It suits users willing to set up a Supabase project and an iOS Shortcut for daily syncing.
Requirements
Runs locally over stdio via the npm package, needing Node 18 or later. Requires a Supabase project you create, with the SUPABASE_URL and SUPABASE_SECRET_KEY environment variables, plus deployed connector and ingest functions. An iPhone with the Health app and an iOS Shortcut for daily sync; loading history uses an Apple Health export.
Before you install
The server holds a Supabase secret key, which grants broad access to your project, so keep it out of shared configs. Setup also mints an ingest key and connector token stored as secrets in your project. The query tool runs SQL against your health database, though it is restricted to read-only statements and a SELECT-only database role. Health data is sensitive; it stays in your own project, but the Shortcut posts it to your project's endpoint.

Installation

In SourceWeft

  1. Open Health Data (your own Supabase) in the dashboard and add it to a workspace.
  2. Enable the server for the chats that should use its tools.

Desktop only via STDIO. STDIO servers start a local process, so they need the SourceWeft desktop host.

Other MCP clients

Follow the launch instructions in the repository.

README

health-mcp

[CI] [npm] [MCP Registry]

Your health data (weight, body composition, calories, macros, water) as an MCP data layer your AI can query. Zero custody: it lives in a Postgres database in a Supabase project you create and own — nobody but you ever holds your data or your keys. An iOS Shortcut syncs it daily from Apple Health; Claude (or any MCP client) reads it back through a connector that runs entirely inside your own project.

Get set up (four steps, ~20 minutes)

1. Create your database + connector

Three ways, pick one — all three end with the same thing: your connector URL, plus your ingest URL and ingest key.

  • The setup page — no terminal, guided clicks through your Supabase dashboard.
  • npx @almostjacked/health-mcp setup — fastest, needs Node ≥ 18. Walks you through logging in to Supabase, picking or creating a project, applying the schema, deploying both functions, and minting your secrets, then prints the connector URL and Shortcut config.
  • Manual dashboard walkthrough — the same steps as the setup page, written out for anyone who'd rather read them first or do them by hand.

Supabase's free tier allows 2 active projects, so this fits alongside one other free project you may already have.

2. Load your history

Without this, adaptive-TDEE needs about two weeks of daily syncs before it has enough data to work. With it, everything works immediately.

  1. On your iPhone: Health app → your profile picture → Export All Health Data. This produces an export.zip (can take a few minutes for a long history).
  2. Get that file to whatever device you're setting up from, then drop it on the setup page's Import panel. Leave Dry run checked first to preview what would be sent before you commit to it.

3. Set up the daily sync

The Shortcut panel has one download button — no signing step, no Terminal. Download the shortcut, install it on your iPhone (iOS asks you to paste in your ingest URL and key during import), then turn on the 9 AM daily automation. Full click-by-click detail (including what to do if a step doesn't survive the iOS import): docs/shortcut.md.

Easiest install (Mac-first): download the signed shortcut on your Mac and double-click it — the two import prompts appear right next to the page's copy buttons, and iCloud syncs the shortcut to your iPhone automatically. (No Mac? Send the file to your phone and answer the prompts there.)

4. Connect Claude

  1. Open claude.ai → Settings → Connectors → Add custom connector.
  2. Paste in your connector URL from step 1.
  3. Save. Claude can now use your health-mcp tools.

If adding the connector fails with "Couldn't register with [name]'s sign-in service", that's a transient claude.ai hiccup, not a problem with your connector — just try adding it again.

Use it

Ask Claude things like:

What's my average calorie intake this week vs last?

Show me my weight trend over the last 90 days.

Has my daily sync stalled? Check get_sync_status.

Pair it with fitness-tools

health-mcp answers "what happened" (your logged weight and calories); fitness-tools answers "so what" (TDEE, macros, body fat, 1RM). The get_energy_inputs tool returns {date, weight, kcal} entries shaped exactly for fitness-tools' adaptive-tdee — no reformatting needed. Ask Claude:

Get my energy inputs for the last 90 days, then compute my adaptive TDEE with fitness-tools.

adaptive-tdee needs at least 10 days that have BOTH a weigh-in and a calorie total — if you skipped step 2, expect ~2 weeks of syncing before the pairing works.

If you also train with Hevy, pair with hevy-mcp for the training side of the picture (routines, logged sets, 1RM trend).

Tools

ToolDescription
get_schemaThe Postgres schema (two tables) and the metric registry (names, units, classes). Call before writing SQL for the query tool.
get_sync_statusLatest date, row count, and days-since-last-entry per metric. Use to detect a stalled daily sync before trusting an analysis.
get_recentRows for the last N days (default 30), optionally one metric. Daily totals and individual measurements in one date-sorted list.
get_daily_summaryAll metrics for one day (default: the most recent day with any data): totals plus every weigh-in/measurement.
get_statsMin/max/avg plus a rolling-average series for one metric over a date range (daily/weekly/monthly rollup).
queryEscape hatch: run one read-only SQL statement (SELECT or WITH…SELECT) against the schema from get_schema. Writes/DDL rejected; LIMIT 500 enforced.
get_energy_inputsDaily (date, weight, kcal) entries for days with both a calorie total and a weigh-in — shaped exactly for fitness-tools' adaptive-tdee.

Architecture

Apple Health --> "Sync Health Data" Shortcut --> your health-ingest function --> your Postgres                                                                                      |Claude  <-- run_readonly (SELECT-only) <-- your health-mcp connector function <------+

Nothing above touches infrastructure we operate. The two Edge Functions run on your Supabase project; the only thing that ever leaves your device is the Shortcut's HTTP POST straight to your own project's URL.

Privacy Policy

Zero custody. health-mcp is software you run against infrastructure you own — we (the maintainers) never operate a server that sees your health data or your keys, and there is nothing to disconnect from us because nothing was ever connected to us.

  • Your data lives only in the Supabase Postgres project you created. We have no access to it, no copy of it, and no way to see it.
  • Your keys (MCP_TOKEN, INGEST_KEY, your Supabase secret key) are generated by you or the wizard running on your machine, stored as secrets in your own Supabase project, and — for the stdio/.mcpb install path — held locally by your MCP client (Claude Desktop / Claude Code) and sent only to your own project's URL.
  • The Shortcut sends data straight from your phone to your Edge Function URL; it never passes through any server of ours.
  • The query tool is SQL-guarded to read-only statements, and the database role it runs as (health_reader) is granted SELECT only — enforced at the Postgres level, not just in application code.

Other clients

Prefer Claude Desktop, or running the server yourself instead of the hosted connector? Both use the same Supabase project from step 1 — just a different way of talking to it.

Claude Desktop: download health-mcp.mcpb, double-click it, and paste in your Supabase project URL and secret key when prompted.

Any stdio MCP client (Claude Code, etc.):

bash
claude mcp add health-mcp -e SUPABASE_URL=<url> -e SUPABASE_SECRET_KEY=<key> -- npx -y @almostjacked/health-mcp

Repo layout

packages/core   @almostjacked/health-mcp-core — tools, ingest/normalization, setup.sqlapps/mcp        @almostjacked/health-mcp       — stdio server, .mcpb, setup wizardweb             the setup page (provision + import + Shortcut panels)supabase/functions/health-mcp     the read connector (deployed to your project)supabase/functions/health-ingest  the write endpoint the Shortcut posts to

Develop

bash
corepack enablepnpm installpnpm -r testpnpm -r typecheck

License

MIT — see LICENSE.

Source: README.md at commit e002ced

Tools

0
SourceWeft has not started this package in its sandbox yet, so its tools are not known.

Version history

1
  1. v0.1.3LatestSep 16, 2026

Score

50% of the applicable checks met

How complete this listing is, computed from the registry entry, the endpoint and public sources. Not a rating of the server and not a statement that it is safe or working today: reachability is a dated check, and no installs or reviews go into the score.

  • Reached in the last 30 days25 pts

    Keep the endpoint answering the MCP handshake; SourceWeft probes it every 30 days.

  • Auth declared10 pts

    Declare the credential headers in the registry entry, or keep the endpoint answering the probe so its auth can be classified.

  • Tools listed15 pts

    Answer tools/list without a credential, or ship a package that starts in a sandbox without one.

  • DescriptionRequired15 pts
  • Logo10 pts
  • Documentation15 pts
  • License10 pts

Other Supabase integrations

Supabase

com.supabaseOfficial publisher

Supabase MCP server that lets an assistant inspect and manage Supabase projects, run SQL and migrations, query logs, and deploy Edge Functions.

Streamable HTTPWeb executableAuth
Databases36 toolsNov 7, 2026

Supabase Mcp

Published by GitHub account mrfentmen

Local MCP server that lets an assistant query a Supabase project's table rows, auth users and RPC endpoints.

STDIODesktop only
DatabasesNot validated yetOct 26, 2026

Supabase URL Shape

Published by GitHub account sadri-dridi

A remote MCP server of small utility checks: URL and hostname shapes, JSON validity, time zones, HTTP status, and a few lookups, with inputs discarded after use

Streamable HTTPWeb executable
Web Search & Scraping31 toolsOct 24, 2026

Supabase

Published by GitHub account mcp-dir

A hosted Supabase connector that lets an assistant inspect projects, run SQL, apply migrations, manage branches and Edge Functions, and search Supabase docs.

Streamable HTTPWeb executable
Databases35 toolsOct 24, 2026

Dataset to Postgres, Supabase & MySQL (Database Push)

Published by GitHub account Nero-Engine

Apify-hosted MCP server that lets an assistant push rows from a dataset, file or Google Sheet into a Postgres, Supabase, Neon or MySQL table.

Streamable HTTPWeb executableAuth
DatabasesCredential requiredOct 24, 2026