See the complete story — not just weight or injections. Record your doses and injection sites, run a 60-second daily check-in, watch an educational medication-level model built from published half-lives, track weight, measurements and habits, then turn all of it into a report you can actually use at your next appointment.
Your estimated medication level, the doses you logged, and the symptom burden you recorded — on one timeline, so you can see how they move together. Association is not causation.
One cell per day from your check-ins. The face and the label carry the meaning; colour only repeats it.
A completeness score for your records. It says nothing about your health, your adherence or how well treatment is going.
Everything except the date is optional. This is a record of what you noticed — it is not a score, a diagnosis or an assessment. You can hide any field you do not want in Settings → Check-in fields.
Once a week, in your own words. Wins, difficulties, questions for your next appointment, and non-scale victories worth remembering.
Record what is on your prescription label, exactly as written. This is a record, not a plan the app has chosen for you.
Every medication record you have kept, newest first. Ending one and starting another keeps your history honest instead of rewriting it.
What you took, when you took it, and where you injected. Nothing here is a recommendation — the app records what you tell it.
Which sites you have used and how long since each one. A record of your own pattern — not an instruction about where to inject.
An educational model, not a blood measurement. Each dose decays independently and the next one lands on whatever has not cleared yet.
A hypothetical schedule you type in, drawn forward so you can look at it with your prescriber. It is not a plan, not a recommendation and not a prediction of how you will feel.
Enter an amount above to see the arithmetic.
| Pen | One click | 2.5 mg | 5 mg | 7.5 mg | 10 mg | Full dial |
|---|
Every figure in this table is pen strength ÷ clicks per dial, rounded for display. Round numbers here do not make a dose correct for you.
Weight is optional, like everything else. If body measurements are not helpful for you right now, skip this whole card — the rest of the app works without it.
The dots are what the scale said. The line is a seven-day moving average, which is much closer to what is actually changing.
BMI is a population statistic, not a description of a person. It ignores muscle, bone, shape and everything about how you feel. You can switch it off permanently and nothing else changes.
Set a target you actually want. Missing days are information, never a failure — there is no streak to "break" here.
The things a scale cannot see. Add one whenever you notice it.
Descriptive associations built only from your own records, and only when there is enough data to say anything at all. Association is not causation — a pattern here is a question to ask, never an explanation.
Untick anything you never want to see. Hidden fields are simply not shown — nothing already recorded is deleted.
Questions you write here appear in your appointment report.
Your records live in this browser's local storage. Clearing site data erases them. Download a backup before you change device or browser — and keep an unencrypted backup file as carefully as you would keep a paper diary.
AES-256-GCM with a key derived from your passphrase (PBKDF2-HMAC-SHA-256, 310,000 iterations, random salt and IV per backup). The passphrase is never stored, never logged and never written into the file. If you forget it the backup cannot be recovered by anyone, including us.
Choose a plain or encrypted backup. Nothing is written until you have seen the preview and chosen Merge or Replace all.
GLP-1 medicines clear slowly. Tirzepatide's elimination half-life is roughly 5 days, semaglutide's roughly 7, dulaglutide's about 4.7, and liraglutide's about 13 hours. A week after a dose of a weekly medicine, about half of it is still in you. The next dose lands on top of that remainder, so the amount in your system keeps climbing for several doses even when the number on the pen never changes.
That one fact explains a lot of lived experience: the same dose feeling stronger after a month, side effects appearing weeks into a stable dose, and appetite returning late in the week. It is also why the app talks about your level rather than your dose — but it never suggests what that level should be.
A one-compartment model with instantaneous input and first-order elimination, expressed as mg-equivalent remaining — not plasma concentration:
Level(t) = Σ Dosei × 2−(t − EventTimei) / HalfLife
Every event carries its own half-life, so changing medicine does not rewrite your history. A date-only record is placed at 12:00 in your time zone. Future events are excluded from history and appear only in an explicitly labelled discussion scenario. If an ingredient has no reviewed half-life value, the app shows no level at all rather than a number that is quietly too low.
Recent doses are almost all that matters. A dose is about 97% cleared after five half-lives, so anything older than roughly 3½ weeks (tirzepatide) or 5 weeks (semaglutide) has essentially left your system. There is no value in reconstructing months of old records.
No records at all? Start with your next dose. Early readings will understate your level rather than overstate it, and the gap closes on its own as unlogged doses clear. Day-level date accuracy is fine — a twelve-hour error moves one dose's contribution by about 7% and your total by far less.
Half-life values are taken from manufacturer prescribing information for each active ingredient and stored with the source text, the basis (population average) and a version tag, so you can see exactly which assumption produced a figure. They are listed beside every chart under Assumptions used.
Everything else on the page is either something you typed or a plain summary of what you typed.
It shows the modelled milligram-equivalent of medication remaining in your body from the doses you entered, using a published population-average half-life for each active ingredient. It is an educational estimate, not a blood measurement, and it cannot tell you whether a level is right or safe for you.
No. It never recommends, calculates, titrates, delays, splits or changes a dose, and it never labels a level or a symptom as safe. It records what you enter and shows you what you recorded. Dose decisions belong with the person who prescribes for you.
No. Because each dose is about 97% cleared after five half-lives, logging your last three to four doses captures roughly 95–98% of your current modelled level for a weekly medicine. Starting from your next dose also works — early readings understate rather than overstate your level, and the gap closes by itself.
In your own browser's local storage on this device. There is no account, no server, no analytics and no third-party embed, and your records are never uploaded. Clearing your browser's site data deletes them, so download a backup before switching device or browser.
It is an arithmetic aid only: it divides a pen strength by the number of clicks in a full dial. It is not advice and not a dosing instruction. Splitting a pen by counting clicks is an off-label technique, so you must confirm every dose against your own pen and your prescriber before you inject.
Yes, under Settings, as an advanced option with an effective date and a place to record where the value came from. Overrides are visually distinguished throughout the app because they change every estimate you see. Most people should leave the reviewed population values alone.
Yes — that is what the Reports area is for. It builds an appointment snapshot or a longer clinician report showing your medication events, symptom summary, trends, questions and a data-completeness note, and it labels every figure as patient-entered, derived or estimated.
Contact emergency services or your clinician straight away. This app does not triage symptoms and will never tell you to wait, log it, or check a chart first.