Home · Community report
Build a personal tracker app
Six people on X report building their own tracker with Muse, from a daily vitamin and medicine log to a fasting tracker tied to Apple Health and a ride-credit app for amusement parks. One builder keeps four apps in daily use but says three other experiments were less successful, so expect iteration.
What you can get
What it is
Personal tracker app
A small working tracker with your existing data imported, and a clear note of what it does not yet do.
It came back right when
- Each entry has the fields you asked for and nothing invented.
- Imported records match the source, checked on a sample.
- No account was connected without your approval.
- You have used it for a week and know what it does not do yet.
The checks are editorial, not from the source.
- Pick one thingStops for your approval
Something you already track badly and think about often.
- Define the fields
What each entry holds and the one view you need.
- Import a sampleStops for your approval
Existing data checked on a small sample first.
- Use and iterateStops for your approval
A week of use, then one change at a time.
Editorial, not from the source
Works when
- You keep thinking about something and have no good place to log it.
- You already have data in a spreadsheet or export that could seed it.
- You are happy to iterate over several sessions.
Does not work when
- You need it right first time with no iteration.
- You would need to hand over logins you are not comfortable sharing.
- You want medical guidance from a medication log. Ask a pharmacist.
Build a personal tracker app: review template
| Item | What it is based on | Status |
|---|---|---|
| [What it tracks] | [Your description and fields] | [Agreed] |
| [Imported data] | [Your export, checked on a sample] | [Checked or pending] |
| [Connected accounts] | [Only those you approved] | [None or listed] |
The source case
On X, @shriyanevatia says they have four apps built with Muse in daily rotation and three other experiments that were not as successful. The four listed are a food and nutrition tracker that also recommends dishes and restaurants, a log of the books, TV, movies and magazines they consume, a simple daily vitamin and medicine tracker, and a clothing app. The media log started from spreadsheets and Goodreads data they imported, and is still filled in by hand. In a separate post about the clothing app, they say it took a ton of hours and tokens of prompting and that they keep iterating. Neither post says what the three unsuccessful experiments were or why they failed. Checked on 2026-09-24, these are individual reports, not evidence of what the reader will get. @chakkaradeep says Muse built both a fasting tracker and conference planner with stated Apple Health and schedule connections; these are two separate attributed app examples, not proof of clinical effectiveness or integration security. @iambchoor reports building bank-linked dashboards for a car, a daily calendar/email feed, investments (via Plaid) and credit-card perks (via Plaid); @dhrandy reports building a bank-linked finance tracker among several other tools; @TheSeanLambert reports a food-and-lift tracker synced two-way with a spreadsheet; and @kbrewFL reports an amusement-park ride-credit tracker with a stats breakdown by type and manufacturer. Two further reports (a Seattle AI-meetup tracker and an unspecified alerts dashboard, already noted above) describe similar builds but add no new detail.
A public report by someone outside this library. Not independently reproduced here.
From the source
Source excerpt
I’m using 4 in daily rotation ... Have 3 other experiments that weren’t as successful ... Log of all media I consume - books tv movies and magazines. Right now it’s manual but I plan to give all streaming service logins. ... Super simple daily vitamin/medicine tracker
Source excerpt
i’ve spent a ton of hours and tokens prompting @Muse to make me this custom fashion app but holy crap is it worth it ... i described everything i like and dislike and i keep iterating each time
Source excerpt
I didn't download a fasting tracker or a conference planner. I described what I needed, and @Muse built them - wired to my Apple Health, my schedule, my targets.
Source excerpt
@Muse Super impressed. I built a dashboard for my car (Tessie), a daily living/breathing dashboard/calendar feed (Gmail/calendar/granola), an investment dashboard (Plaid), a deals review skill and was able to dive deep into 3 deals, and a credit-card perks dashboard (Plaid).
Source excerpt
@Muse Hardest things my AI assistant has built for me: a bank-linked finance tracker, a self-hosted email relay, a 3D-printer monitor with finish detection, a secure bridge for my media server, and a full rebuild of my vinyl collection app.
Source excerpt
@Joey_vm_Ware @aaronbuley @Muse @home_assistant @MyFitnessPal I built a food tracker and lift tracking app to sync with an existing master spreadsheet that syncs both ways…
Source excerpt
I had my Muse agent throw together an app for tracking your water ride credits! The catalog is still being built out, but the UI is functional. You can add rides you’ve been on and rate how soaked you got on them. And there’s a fun stats breakdown by type and manufacturer! https://t.co/lzV1lnLlw2
Quoted from the original posts. Not written by this guide, and not a result produced here.
Suggested prompt
Bring What you want to track, The fields for each entry, Existing data as an export or spreadsheet, How you want to see it. Acts Builds a small app from your description. Imports data you provide after you check a sample. Does not connect new accounts or logins without your explicit approval. Source community report, checked 2026-10-09.
Written by this guide as a starting point, not quoted from the source. Your edits stay in this browser.
Other ways to ask
Editorial rewrites of the brief above, for a different emphasis. Copy them separately.
What you need
Editorial, not from the source
You supply
- What you want to track
- The fields for each entry
- Existing data as an export or spreadsheet
- How you want to see it
Connected
- Only the data sources you choose to share
Approved by you
- Your approval before any account or login is connected
- Your choice of what personal data goes in
- Your check of a sample import
How to try it
Editorial, not from the source
- Pick one thing you already track badly, such as a list in your notes or a half-kept spreadsheet.
- Write down the fields each entry needs and the one view you will actually look at.
- Hand over existing data as an export or spreadsheet and check a sample import before the rest goes in.
- Use it daily for a week before adding features, and note what you keep wanting.
- Decide what personal data goes in. Photos, measurements and health details deserve a deliberate choice.
Start with what you already think about
The author of the report describes building apps for anything they constantly think about: food, the media they consume, daily vitamins and medicine, and clothes. That is a useful filter. A tracker is worth building when you already reach for a notes list or a spreadsheet and keep abandoning it. Before asking for anything, write down what each entry needs, such as a date, a name and a rating, and the one view you will look at most. Resist the urge to describe everything at once. Four of the author's seven experiments made it into daily use, and the other three were not as successful. A small first version that you open every day tells you more than a large one you admire once and forget.
Bring your old data, carefully
In the report, the media log started from spreadsheets and Goodreads data the author already had, imported to form the core of it. Existing data is the fastest way to make a new tracker feel useful on day one. Hand it over as an export or a file rather than a login where you can, and ask to see a small sample after import before the rest goes in. Check dates, titles and anything with numbers. The author says the log is still manual and that giving it streaming service logins is a plan, not yet done. That is a reasonable pause. Each account you connect is access you are granting, and it is worth deciding whether the automation is worth it before you do.
Expect to iterate, and to drop some
The clothing app in the second post took, by the author's account, a ton of hours and tokens of prompting. They gave it their height, weight, measurements and photos, their favorite stores and budget ranges, and keep refining it by describing what they like and dislike. That is the honest shape of a custom app: the first version is a draft. Change one thing at a time and say specifically what is wrong. Think too about what you are putting in. Photos and body measurements are personal, and a vitamin or medicine tracker holds health information; it records what you enter and does not check doses, which is a pharmacist's job. If an idea is not working after a few rounds, let it go.
Open Muse to try it. That is the product's own site; this guide is not part of it.
Limits and confirmation points
Limits to confirm
- Seven posts from six people. Only one mentions failures, three experiments that were not as successful, without saying why; the rest list what was built, not how well it works.
- By the author's account, the clothing app took many hours of prompting and repeated iteration. Expect a first version to need work.
- The media log is still manual in the report. Connecting streaming logins is described as a plan, not something done.
- A vitamin or medicine tracker records what you enter. It does not check doses or interactions; ask a pharmacist or clinician.
- No source code, Apple Health permission scope, actual fasting records, event data or test evidence is provided.
- Several reports connect a bank account or provider (e.g., Plaid) to a self-built tracker; review what access a custom app is requesting before approving it.
Stop and check
Editorial, not from the source
- Before building: agree the fields and the one view you need.
- Before importing: check a sample of your existing data came across correctly.
- Before connecting anything: approve each account or login explicitly.
If it goes wrong
Editorial, not from the source
- The first version is not what you pictured
- Expect that. In the report, the clothing app took many rounds of prompting. Describe what you like and dislike specifically, one change at a time.
- An idea does not stick
- Let it go. The same author reports three experiments that were not as successful alongside four in daily use.
- Entries still need typing in by hand
- That is where the media log in the report still stands. Decide whether connecting another account is worth the access it needs before asking for it.
Common questions
Can Muse build a custom tracker app?
One user reports four apps built with it in daily use, including a vitamin and medicine tracker and a media log, and three experiments that were not as successful. This directory has not reproduced that.
How long does it take?
It varies. The same author says the clothing app took a ton of hours and tokens of prompting and is still being iterated. Start small.
Can I import my existing data?
In the report, spreadsheets and Goodreads data were imported to seed the media log. Check a sample before importing everything.
Can it track my medication?
It can record what you enter, as in the report's simple vitamin and medicine tracker. It does not check doses or interactions; ask a pharmacist or clinician.
Do I need to give it my logins?
Not to start. In the report, connecting streaming logins is a plan rather than something done. Decide account by account.
Next
Related to this task
More in Home
Similar tasks
Get recurring task recommendations — ask your own agent to check for relevant updates.