PDF, Word, Excel, PowerPoint and ten more formats, converted on your own hardware. Tables, footnotes and math survive the trip. Nothing is uploaded — there is no server to upload to.
Coming to the App Store. Pick a platform and we will write once, on launch day.
Headings stay headings. Lists keep their nesting. Tables keep their columns. Footnotes keep their anchors. This is a real conversion, not text scraped off a page.
No account, no server, no network call. The conversion happens inside the app on your device — the only arrangement that works when your documents are contracts, case files or patient records.
Merged cells keep their shape, footnotes keep their anchors, formulas come out as LaTeX — so a report still reads like a report after conversion, not like debris.
Every file gets a per-page report. You never have to read the whole output just to find out whether a page went quietly missing.
Scanned pages are recognised on-device with Apple Vision, and only the pages that actually need it: a 200-page file with three scans costs three pages of battery, not two hundred.
DocNavy is for the documents where the file is long, the table matters, or the content cannot leave the building.
Books, papers and reports have to land in Obsidian or Logseq as real Markdown — headings intact, folder structure mirrored, images next to the note. Not a wall of text you have to re-format by hand.
Chatbots read Markdown far better than a pasted PDF. The difference shows up in tables: a price list that survives conversion is a price list the model can actually reason about.
Law, accounting, healthcare, finance. “Do not send client files to a third party” is not a preference, it is the rule you work under. DocNavy has no server to send them to — and macOS blocks it from opening a connection at all.
Cases from our own regression suite — 14 formats × 4 difficulty levels, run on iPhone and Mac before every release.
The engine classifies every page first, so OCR runs on the three that need it — not on all two hundred. In the suite the recovered pages come back with their headings intact (# and ##), not as flat text.
A 6,000-row sheet converts with every row present — the assertion fails the build if a single one goes missing. Merged cells keep their span, and text boxes anchored mid-table land between the right rows, not dumped at the end.
The classic failure here is silent: the text extracts as garbage and nobody notices until much later. DocNavy ships the character maps inside the app and the suite checks a CJK-embedded PDF on every build.
A SmartArt tree is compared against a stored reference — parent-child relationships and drawing order both have to match. Formulas are checked the same way, as block $$…$$ and inline $…$.
Two minutes with a tool Apple ships in every iPhone.
Open Settings › Privacy & Security › App Privacy Report and turn it on.
Use DocNavy normally for a few days — convert whatever you like.
Open DocNavy’s entry. The list of domains contacted should be empty.
On the Mac the app ships without the network entitlement at all — macOS will not let it open a connection even if it wanted to. That is enforced by the operating system, not by our good intentions.
Not a crippled demo. Every feature is unlocked from the first launch — paying only lifts two limits.
Identical in every tier: no upload, no account, no watermark, no telemetry, full output quality. Paying changes how much you can convert — never how well.
No. There is no sign-up, no login and no password to lose. Your purchase rides on your Apple ID, which is also how it reaches your other devices.
It converts the first 50 pages at full quality and tells you how many remain. If you buy Pro afterwards it continues from page 51 — pages already recognised by OCR are not processed again.
Monthly if you have a burst of work and want out afterwards. One-time if you expect to keep using it: at $2.99/month the subscription passes $29.99 in the tenth month, and after that the one-time purchase is simply cheaper.
Into a folder you choose on your own device. Nothing is copied anywhere else, and deleting the app deletes its working files with it.
Always. Offline is the only mode there is — including OCR, which uses the recognition engine built into iOS and macOS.
That is exactly who it was built for. Folder structure is mirrored in the output, images are extracted next to the Markdown, and headings, tables and links land in a shape a vault can use.
iPhone and iPad on iOS/iPadOS 17 or later, and Mac on macOS 14 or later.
Already using DocNavy and something went wrong? The support page covers troubleshooting — garbled pages, OCR languages, restoring a purchase.
Coming to the App Store for iPhone, iPad and Mac.
Requires iOS/iPadOS 17 or macOS 14.