Most sites make this claim in a policy document. Here it's a property of how the site is built — and you can verify it yourself in about thirty seconds.
Open any tool and press F12 for developer tools
Switch to the Network tab
Process any file and watch the requests
Requests carrying your file: zero
Concrete measures, not reassurance.
Every library and font is served from this site's own folder. No CDN, no Google Fonts, no analytics, no QR service. A supply-chain attack on a CDN cannot reach you, because nothing is fetched from one. The only outbound call anywhere on the site is the contact form delivering your typed message — and it is blocked on every other page.
default-src 'self' with object-src 'none', base-uri 'none', form-action 'none' and frame-ancestors 'none'. Injected external scripts, clickjacking frames and form hijacking are all blocked.
This is a static site. There is no backend to breach, no database to leak, no accounts to steal and no upload endpoint to attack. Entire categories of vulnerability simply do not apply.
Only six libraries ship, each pinned to an exact version and vendored locally. Nothing updates itself behind your back — the polyfill.io style attack is impossible here.
A no-referrer policy is set site-wide. There are no advertising cookies, no analytics scripts and no accounts. The only thing stored locally is nothing at all.
Plain HTML, CSS and JavaScript. Press Ctrl+U and read every line, or open the folder and check the files. There is no build step hiding what actually runs.
A privacy page that only lists strengths is marketing. Here are the real boundaries.
Only on the contact page, the message you type is passed to EmailJS so it reaches our inbox. It carries just what you wrote — never a document, never anything from the tools. Every other page is blocked from contacting them at all, and you can always copy the text and email us yourself instead.
Because everything is local, the safety of your files depends on your own machine. If your device has malware, no website can protect your documents from it.
Whoever hosts these files can see that a browser downloaded the page, as with any website. They never see your documents — those are never sent.
OCR needs a 15 MB language model that would have to come from an external CDN. Rather than break the no-third-party rule, it ships disabled. The README explains how to enable it locally.
Real PDF encryption cannot be done safely in a browser. We mark that tool unavailable rather than ship a fake lock that gives false confidence.
Removing the server and the CDN doesn't make a site magically secure — it removes entire categories of attack that cannot happen when there's nowhere for your data to sit and nothing being fetched from a third party. What remains is protecting the hosting account itself, which is standard practice: strong passwords and two-factor authentication.