Anonymise Word, Excel and PDF documents without them ever leaving your computer.
Drag a file onto the page, choose what to hide (names, tax codes, IBANs, addresses…), download the same file cleaned. Everything happens in your browser: zero uploads, zero servers, zero cloud. A single HTML file to save and use offline.
The tools for anonymising are all wrong.
Every week someone in the company shares a file with a supplier, a consultant, an agency. A CV, a sales export, a contract, a medical report. The options today aren’t good.
Redacting by hand in Word/Excel
Find & replace leaves traces: history, metadata, copies, tracked changes. And it doesn’t scale beyond a few files.
A black box on a PDF
In Acrobat, without flattening, the text can still be extracted with a copy-paste. The most common mistake in law firms and healthcare.
Uploading it to a cloud tool
ChatGPT, a free PDF redactor, an online service. A GDPR breach: personal data ends up on servers you know nothing about.
Three steps, nothing to install.
Drag your files onto the page
One or more Word, Excel or PDF documents. The tool runs in the browser: files stay on your PC and are never uploaded anywhere.
Choose what to hide
First names, surnames, tax codes, IBANs, addresses, emails, phone numbers. Fine-grained selection by data type, with a preview before applying.
Download the cleaned file
Original format intact: a .docx stays a .docx, an .xlsx stays an .xlsx. With an optional reversible pseudonymisation if you need to read them back later.
Built for real Italian data, not for the generic Anglo-Saxon case.
Generic tools detect the Italian tax code at roughly 67% (they stop at the 16-character pattern and get the homocode variants wrong). Our detector is tuned on real Italian patterns.
A 2 MB HTML file. No installation, works offline.
Just leave us your details (we use them to send you updates and ask for feedback in a week). Zero spam, promise.
Frequently asked questions
Are my files really not uploaded anywhere?
Yes, really. All the detection and anonymisation logic runs in the browser via WebAssembly. If you switch off the Wi-Fi after opening the tool, it keeps working. Check for yourself with the browser developer tools (Network tab): no request with your files ever goes out.
Which formats does it support?
Word (.docx), Excel (.xlsx) and native PDFs (those with selectable text). It doesn’t run OCR on scans: if you have a scanned PDF, OCR it first with another tool. Outside the MVP: PowerPoint and legacy formats (.doc, .xls).
Which anonymisation techniques does it apply?
Four: (1) redaction — covers the data with a box, (2) replacement with consistent fake data (Italian Faker — Mario Rossi becomes Luca Bianchi but stays a plausible name), (3) pseudonymisation with a local vault — generates a stable ID and keeps a separate key you can save to restore the real data later, (4) generalisation — e.g. 12 March 1985 becomes "1985" or "1980s".
Does it detect the Italian tax code better than generic tools?
Yes, and it’s one of the reasons we built it. Generic NLP tools (Presidio, spaCy) stop at the 16-character regex pattern: they miss homocodes and reach ~67% accuracy. Our detector adds check-digit validation and composition rules, taking accuracy above 95% on real Italian cases.
Do I need to install anything?
No. You download a single HTML file of about 2 MB, save it wherever you like, double-click and it opens in your browser. No installer, no account, no background service. You can even put it on a USB stick and use it elsewhere.
Is it open source?
The frontend and the detection engine are open source (AGPL-3.0). Backend, enterprise features and billing remain proprietary. Public repo coming soon — if you want to be notified when we publish it, tick "contact me" in the form.
How much does it cost?
The tool is free and will stay free, with all MVP features. We’re considering a Pro plan (planned for phase 2) with advanced features such as batches of 100+ files, an API, and reversibility with an encrypted vault. If you want lifetime updates whenever we publish new versions, tick the dedicated option in the form (one-off contribution of €29).
Can I use it to comply with the GDPR and the AI Act?
It’s a practical tool that helps you respect the data minimisation principle (GDPR art. 5) and good pseudonymisation practice (EDPB Guidelines 01/2025). On its own it isn’t enough for compliance: compliance is a process. But it’s exactly the missing piece of workflow we see in almost every Italian company. If you want support on the overall picture, tick "contact me about the AI Act" in the form.
A tool built to be used.
If you help us improve it, it improves for everyone. Download it, try it on real files, and if you can, reply to the feedback email you’ll get in a week.
