Press

Everything on this page is checkable. If you are writing about Scanly and something here does not hold up, that is a bug and we would like to hear about it: help@deplyra.com.

In one sentence

Scanly is a set of document tools — scan to PDF, JPG to PDF, PDF to JPG, compress PDF, and an ID card layout — that run inside the browser, so the file never reaches a server.

The claim, and how it is enforced

Your document never leaves your browser. That is the product, and it is held up in four places rather than asserted once:

• There is no endpoint. The site is static files. The server answers 405 Method Not Allowed to any request carrying a body, and a check on every deploy fails if that stops being true.
• The code cannot send one. Each tool's engine is scanned at build time for any way of making a network request. If one appears, the site does not build.
• The browser is instructed to refuse. A Content-Security-Policy forbids form submission outright.
• It is watched. Every tool is driven with real files in a real browser on every deploy, and every outgoing request body is inspected. A run in which the document appears in one fails.

You can verify the first claim yourself in about ten seconds: open the developer tools on any tool page, switch to the Network tab, and run a scan. Nothing goes out.

What it does not do

• No text recognition on the web. The PDFs hold pictures of your pages. Making a scan searchable is in the Android app, and that recognition reads Latin script only.
• It is not a replacement for a flatbed scanner. For archival photographs a real scanner wins clearly, and the site says so on the page where someone would otherwise assume otherwise.
• No accounts, and nothing stored. There is nothing to sign into and no history, because there is nowhere to keep one.

Who makes it

Deplyra Ltd, registered in England & Wales. Scanly also exists as an Android app, currently in closed testing with no public listing.

Facts you can quote

• Five tools, all free, no sign-in, no daily cap, no watermark.
• They are free because they run on the reader's device and cost nothing per use — not as an introductory offer.
• ID cards are placed at 85.6 × 54 mm, so a print at 100% measures the same as the card.
• The edge detector is about seven kilobytes, written for this. The usual approach is a ten-megabyte computer-vision library, which every visitor would have had to download.
• Urdu versions of the main tool pages are published in Nastaliq, not Naskh.

What we will not say

The build carries a list of marketing phrases it refuses to ship — the security-sounding ones that mean nothing on inspection, the accuracy claims no text recognition can honour, and the fashionable ones describing capabilities this site does not have. A page containing one does not build. It also refuses any star rating, because there are no reviews to base one on.

That list is not printed here, and the reason is worth stating: this page is subject to the same rule as every other. Quoting the banned phrases in order to say we ban them would either break the build or require an exemption for this page, and a law with an exemption is a preference. The list is in scripts/build.py. If you see a claim of that kind attached to Scanly anywhere, it did not come from us.