Marrow

Marrow Health: your Apple Health data over MCP, from the free iOS app.

Connector by skeletonarmy.tech · Domain verified by the MCP Registry

IndexedMITRemote (HTTP)★ 3

Write a review

Install

Pick your client. Read the repository before you give any tool access to your data.

No install instructions from the source yet. Check the repository for setup steps.

Get the weekly diff

The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.

About

Marrow Health: your Apple Health data over MCP, from the free iOS app. Reads, and writes on demand.

Reviews

Write a review

No reviews yet. Installed it? Reviews come from verified installers only. How reviews work

Permissions

What this server can do and what it asks for, from its tool list, install config and code. Checked Oct 25, 2026.

Tools: 3 read · 9 write · 2 delete · 13 not yet classified
Secrets: None asked for
Install scripts: None found
Runs as: Python package

Tools by what they do 27

Read from the source code; the running server may list different tools.

Read 3 · Look things up without changing anything.

health_summary
list_metrics
metric_samples

Write 9 · Create, change, send or run things.

edit_entry
log_meal
log_mood
log_sleep
log_symptom
log_water
log_weight
log_workout
set_targets

Delete 2 · Delete, drop, revoke or wipe things.

delete_entry
log_strength

Not yet classified 13 · We could not tell from the name and description.

baselines
daily_brief
data_quality
metric_daily
nutrition_log
pending_entries
sleep_log
strength_log
symptoms_and_mood
targets
undo_entry
what_changed

Show 1 more tools

workouts

Network hosts it talks to 5

connect.marrowhealthapp.com · found in code
github.com · found in code
glama.ai · found in code
marrowhealthapp.com · found in code
pi.local · found in code

Tools (27)

Read from the source code; the running server may differ. Descriptions are the server’s own words; hidden characters are shown as [U+…].

baselines
The user's own normal per metric: mean and spread over the trailing window, the latest value's z and phrase, the 7-day mean and the 90-day trend per month. The same math as the app's Home page. Use it to say whether a number is high or low for this person instead of against population figures.
daily_brief
One day in the app's own terms: each key metric with its value and a phrase against the user's own normal (normal for you, above your normal, well below your normal), food and water against the targets, last night's sleep, training done, mood and symptoms, and flags worth mentioning. Start here for how am I doing, what should I know today, or before giving any advice. Default day is the phone's today.
data_quality
What the mirror actually holds and how fresh it is: per metric the latest day, coverage over the recent window, cadence (daily or occasional), its sources and whether it has gone stale, plus diary, sleep and training coverage, when the phone last synced and whether write-back is on. Call it before trusting an absence: no sleep in the mirror is not the same as no sleep.
delete_entry
Remove one food diary entry (from Marrow and Apple Health) by its id from nutrition_log, the user's own or one you added. Only when the user asks. undo_entry with the returned id restores it.
edit_entry
Change one food diary entry: its portion, its numbers or its meal. Works on any entry in nutrition_log, the user's own or one you added, by its id. Give scale to multiply the whole entry (0.5 = half), or grams to set a new portion (calories and macros follow in proportion unless you also give them), or the numbers to set outright. undo_entry with the returned id puts the entry back as it was.
health_summary
The user's recent days as a plain table, one value per day: steps, active calories, exercise minutes, resting heart rate, heart rate variability, body weight, calories eaten and hours asleep. For "how am I doing" prefer daily_brief, which scores each value against the user's own normal.
list_metrics
Every Apple Health metric synced from the user's iPhone, with its unit and the days it covers. Pass a 'key' from here to metric_daily, metric_samples or baselines.
log_meal
Add foods to the user's Marrow food diary. Marrow on their phone applies the entry (labelled as added by you) and writes it to Apple Health. One call per meal. Every item needs calories and macros for the portion actually eaten: use label or database values when you have them, otherwise your best estimate, and tell the user which. Check nutrition_log first when they say 'the same as yesterday'. Returns an id that undo_entry accepts.
log_mood
Record how the user feels in Apple Health's State of Mind (needs iOS 18): a momentary emotion or the mood of the day, on Apple's scale, with the feelings and what it is about when they say.
log_sleep
Record a night (or nap) the watch missed: when the user fell asleep and woke, phone time. Refused when that time already has sleep recorded (check sleep_log).
log_strength
Record a weightlifting session with its sets: every exercise, reps and weight, and a kind when a set is a warm-up, a failure set or a drop set. Sets live in Marrow's strength log (strength_log shows them). Apple Health gets the workout. Use the exercise names the user already has in strength_log where they match.
log_symptom
Record a symptom the user reports in Apple Health, with how bad it is. Only what they tell you. Never infer a symptom.
log_water
Add water the user drank (ml or oz) to Marrow and Apple Health.
log_weight
Record the user's body weight (kg or lb) in Marrow and Apple Health, with body fat percent if they gave it.
log_workout
Record a workout the user did but did not track: a run, a ride, a class. Goes to Apple Health as a workout. For weightlifting with sets and reps use log_strength instead.

Show 12 more tools

metric_daily
One metric as one value per day (a day's total for steps or calories, its average for heart rate and the like). Prefer this to metric_samples for trends and comparisons between days.
metric_samples
The individual records behind one metric, each with its time, value and source app or device, newest first. Use it when the times of single readings matter. For daily totals use metric_daily. When a page is full (limit records), pass the last record's start as before to get the next page.
nutrition_log
The meal diary: every logged food with full nutrient detail, grouped by day. Entries an AI added that the phone has not applied yet are included and marked pending.
pending_entries
What you sent in the last 48 hours and what became of each: queued (waiting for the phone), applied, failed, cancelled or undone.
set_targets
Change the user's daily targets in Marrow: calories, macros, water, steps, sleep, exercise minutes. Only the ones you pass change. Calorie and macro targets switch to custom (the app stops recomputing them). Read targets first, and say what you changed.
sleep_log
Nights of sleep, one per wake day, exactly as the app and Apple Health count them: bedtime, wake time, hours asleep, and minutes per stage (core, deep, rem, awake). Use this for any sleep question instead of raw metric_samples, which cannot be summed by calendar day.
strength_log
Weightlifting detail: every set (exercise, weight, reps, set_type: work, warmup, failure or drop) with estimated 1RM, per-exercise progression, and volume per muscle group. Warm-up sets are listed but left out of progression, volume and muscle credits, as in the app.
symptoms_and_mood
Symptoms logged in the Health app with their severity in words, and State of Mind entries (daily mood or momentary emotion) with valence, feelings and what they were about. Use this for questions about symptoms or mood the user logged in Apple Health or Marrow.
targets
The user's daily targets as set in Marrow: calories, macros, water, steps, sleep and exercise minutes.
undo_entry
Take back something you did, by the id its tool returned. Still waiting for the phone: cancelled. Already applied: a meal, workout, sleep, symptom or mood entry is removed from Marrow and Apple Health. an edit, a delete or a targets change is put back as it was. Water and weight already on the phone can only be removed in the app.
what_changed
The recent days against the period before them: per metric the recent and prior means, the change and how notable it is for this person, plus food, training, sleep and bedtime shifts, and any stream that started or stopped arriving. Notable items come first. Use it for weekly reviews and for what is different lately.
workouts
The user's workouts from Apple Health, from every app and device: activity, start, end and duration.

Signals

Stars: 3
Forks: 1
Open issues: 0
Weekly downloads: 684
Last commit: Oct 23, 2026 (6 days ago)
Last release: Oct 22, 2026 (7 days ago)
Last package publish: Oct 22, 2026 (7 days ago)

From GitHub (fetched Oct 24, 2026) and the package registry.

Details

Registry name: tech.skeletonarmy/marrow
Version: 0.4.1
Auth: No auth
Packages: pypi marrow-mcp
License: MIT
Repository: github.com/skeletonarmytech/marrow-mcp
Website: marrowhealthapp.com/mcp
First seen: Oct 24, 2026
Last checked: Oct 24, 2026
Source: Official MCP Registry github-graphql

Cite this page

Marrow MCP server. Agent Mart. https://agentmart.dev/connectors/skeletonarmy-tech-marrow (last updated 2026-10-24)