Privacy
Every statement below maps to the actual data flow. Verify it yourself in your browser's network panel.
1. File processing happens in your browser
All PDF tools run engines inside your browser (pdf.js / pdf-lib / tesseract.js) directly on local files. Your files are never sent to any server — the network panel only shows page assets (HTML/CSS/JS), never an upload.
2. No files on any server
This site has no file-upload endpoint and no server-side file storage. Not "stored then deleted" — architecturally unable to receive your files.
3. AI PDF assistant (the one exception)
The AI assistant must send document text to a model. Flow: text extracted locally → sent anonymously with your question to Cloudflare Workers AI → streamed back. We store no conversations and there is no account system. If you prefer zero data leaving your device, simply don't use the AI assistant — every other tool has no such data flow.
4. Local storage
localStorage holds two items: the daily run counter (date + count) and an anonymous device ID — used for the free quota and the optional task history view. History stores metadata only (tool name, file name, size, time), never file contents or identity; clear it anytime from the history page or by clearing browser data.
5. No accounts, no tracking
No signup, no analytics, no tracking cookies.
6. OCR language models
First OCR use downloads a language model once (5–20MB) from a public CDN. That's a download, not an upload — recognition still happens on your device.
7. Static assets & CDN
The site is served by Cloudflare's network, which logs standard request metadata (URL, time, IP) for security and operations — the same as every website, unrelated to your file contents.