MyTailLog

Aircraft logbook digitizer & maintenance tracker

Your annual is in three weeks. The logbooks are a cardboard box.

Photograph the pages, or upload the scans you already have. A vision model reads each one into a structured entry — the date 1979-04-12, tach 2214.7, the work performed, the parts, AD 79-08-03, the signing mechanic and certificate number — scores its own confidence field by field, and puts the result beside the original page for you to confirm.

Minutes per page plus your review time, and nobody else ever handles your books. The other way to get decades of paper into data is to mail them to a transcription service, wait days to weeks, and pay in the hundreds per aircraft.

Email link or a password — one page, no card, nothing to cancel. Every new account already has a demo aircraft in it (N734DM, a fictional 1978 Cessna 172N with decades of history), so there is something to poke at before you scan anything of your own.

The review screen: a handwritten 1979 logbook page on the left, the extracted entry on the right with a confidence percentage beside each field and the uncertain ones flagged
Capture & readThe page as written, beside what the model read — with a confidence score on every field and the shaky ones flagged for you.

What's due

Then it starts telling you things

Every inspection, recurring item and AD on one grid, coloured by how close it is. Part 91 recurring items are tracked on hours andon the calendar, and an hours countdown is projected onto a date from how much you actually fly — so “due in 38.4hours” also reads as a month you can plan around. Every projection is marked , carries a confidence drawn from how much reading history backs it, and never overrides a real calendar limit.

AD and SB compliance carries an official FAA reference lookup — the Federal Register for anything post-1994, with a DRS fallback for the legacy ones — and AD discovery searches by manufacturer, model or keyword, listing the models each AD names with yours highlighted.

Email reminders go out before things come due, with a lead time you set per item.

The Status grid: inspections, maintenance items and airworthiness directives listed by urgency, each with hours and dates remaining
Status — everything due, ordered by how close it is.

Ask it

Questions, not file names

Ask in plain English — when was the vacuum pump last replaced, has the seat rail AD ever been done — and the answer arrives with the entries it came from cited underneath, so you can check the reasoning against the scan rather than trust it.

The same reading builds an installed-equipment list from what the logs say went in and came out, and a records-gap audit that points at the stretches where the book simply skips.

Ask your logbook: a plain-English question answered in a paragraph, with the source log entries listed below it
Ask — answers that cite the entries they came from.

Find it

One timeline across every book

Airframe, engine, prop, avionics and a catch-all “other” book, merged into a single searchable timeline. Search the work text, filter by book, jump to the scan the entry was read from.

Duplicate detection flags likely repeats by date, tach/hobbs and work text, so re-captures and re-extractions don't quietly pile up — and hobbs and tach readings are reconciled against each other rather than averaged into a number that flatters your hours.

Timeline and search: log entries from every logbook merged into one chronological list with a search field and per-book filters
Timeline — every book, one chronology, searchable.

Keep it

The documents, not just the entries

Weight & balance is kept as a revision history with the current empty weight and CG on top, flagged stale when equipment changed after the last revision. Scan a new W&B sheet and it files itself as the next revision; scan an A&P's AD compliance report and it becomes the ground truth for your AD state.

The Records Vault holds the permanent paperwork beside the logbook scans — airworthiness certificate, registration, radio station authorization, POH/AFM, STCs, 337s, 8130-3s, manuals.

Weight and balance: current empty weight, CG and useful load above a list of dated W&B revisions
Weight & balance — current figures over the full revision history.

Per aircraft

And one place it all lands

Each aircraft gets its own view: status, hours, open squawks, oil trend, what's next. Squawks are reported by anyone with access — including a pilot you've shared the aircraft with — carry a severity and a reporter, and stay open until an editor resolves them.

Share an aircraft as viewer or editor, transfer ownership, or delete it. Everything is scoped by Postgres row-level security rather than application code, through a single access check every table routes through.

Aircraft overview: hours, next due items, open squawks and recent activity for a single aircraft
One overview per aircraft, colour-coded by what needs attention.

Why it is free

Free is a fact about the code, not a promise about next year

There is no billing in MyTailLog at all — no plans, no tiers, no trial clock, no card field, no payment code anywhere in the app. The repository is public and MIT-licensed, so that is not a claim you have to take on trust; it is a thing you can go and check.

It also means the exit is real. Everything you put in comes back out as a .zip of every row as JSON plus the original scans, re-importable into any instance — including one you run yourself, on your own Supabase and your own Anthropic or OpenAI key. Portability that survives us losing interest is a stronger promise than portability that depends on us staying nice.

The one line item that costs real money is the AI. Reading a logbook page is a vision-model call, billed to whoever's key is used — so shared use is capped two ways, a per-person daily call limit and an overall daily-dollar ceiling. If the day's shared budget runs out, extraction pauses until tomorrow and everything else keeps working.

Add your own Anthropic or OpenAI key in Profile and the calls bill to your account at list prices, with a much higher limit. The numbers, and everything else about cost and privacy, are in the FAQ.

The rest of it

Everything else it does

Capture & extract

  • Capture with your phone's own camera — or upload PDF, JPEG, PNG
  • Pages queue on the device with no signal and upload when you're back
  • Five logbook types: airframe, engine, prop, avionics, other
  • Per-field confidence scores; low-confidence fields are held back from bulk-confirm
  • CSV import — one AI pass maps your columns, then every row converts in plain code. CSV only, up to 5 MB / 5,000 rows, maintenance entries
  • Duplicate finder across scans and entries
  • Scan a W&B sheet or an AD compliance report and it applies itself

Track & forecast

  • Status grid, coloured by urgency
  • Part 91 recurring items on hours and on the calendar, with an projected date and a confidence
  • AD/SB compliance with FAA reference lookup (Federal Register, DRS fallback)
  • AD discovery by manufacturer, model or keyword
  • Installed equipment rebuilt from the logs
  • Weight & balance revisions, flagged stale after an equipment change
  • Records gap audit
  • Squawks — anyone with access reports, an editor resolves
  • Email reminders with your own lead time per item

Hours & engine health

  • Hobbs↔tach reconciliation, so an inflated hobbs doesn't drive your forecast
  • MyFlightBook sync pulls your latest hobbs/tach once a day
  • ADS-B passive hours — opt-in per aircraft, off by default; it only ever suggests a meter value, never writes one
  • Oil analysis — a lab report in, wear metals charted against the lab's universal average
  • Oil consumption — hours per quart between top-offs
  • Bring your own Anthropic or OpenAI API key to bill AI usage to your own account

Keep it & leave with it

  • Records Vault for the permanent paperwork
  • A printable maintenance summary — the page you hand a buyer, an insurer or an IA
  • A full .zip backup of every row as JSON plus the original scans, re-importable
  • Automatic backups to your own Dropbox and/or Google Drive, monthly or quarterly
  • Sharing (viewer / editor), ownership transfer, delete
  • A read-only OAuth 2.1 API and a self-serve developer portal
  • An iOS app in TestFlight beta that browses everything offline
  • MIT licence — run the whole thing yourself

Limits

What it doesn't do

No accuracy percentage, because nothing here measures one
What you get instead is a confidence score per field, the scan beside the entry, and the flagged fields held back from bulk-confirm. That catches what the model knows it is unsure of. It cannot catch what the model was confidently wrong about — your review is what closes that gap, and it is real work.
No scheduling, dispatch, invoicing or billing
Deliberately out of scope, not a roadmap item. Same for eSignatures, parts inventory, work orders and multi-fleet MRO management. If you need a booking calendar and rental rates, an ops platform does that and this does not.
Mobile is a beta, and iOS only
The native app is in TestFlight. It browses every entry, document and scan fully offline and captures pages offline, but editing existing entries is still done on the web, and there is no Android app.
CSV import is CSV, and it is maintenance entries
Not XLSX — save it as CSV first, which is one step. Up to 5 MB and 5,000 rows, creating log entries only, and everything imported lands unconfirmed until you have reviewed it.
This is an index, not the record. The physical logbooks remain the system of record per 14 CFR 91.417. MyTailLog does not replace official maintenance records, is not an airworthiness determination, and is not a maintenance sign-off. Every value shown is derived from OCR/extraction — confirm it against the paper logbook before you rely on it.

Start with one page and see what comes back

Sign in with an email link or a password. Nothing to install, no card, and a demo aircraft is already loaded so you can look around first.