تنقيح PDF.

20 MiB, 200 pages and 2,000 boxes. Search uses the embedded text layer: scanned pages need manual boxes or OCR PDF first. Redacted pages become images (110–300 DPI), so their remaining text is not selectable; links, form fields and comments on those pages are flattened.

Every page that has a box is rendered in your browser, the boxes are painted into the pixels and the page is rebuilt as a single image, so the covered text, images and vectors are deleted rather than hidden. Pages without boxes are copied unchanged. Title, author, subject and XMP metadata are removed by default; bookmarks, attachments and digital signatures are never carried over. After export the file is re-read: the tool confirms that no selectable text is left on redacted pages and lists any page where your search terms still appear.

Remove a name, an email address and an IBAN from a two-page statement before sharing it.