InkedOut removes text and images from a PDF instead of covering them, entirely in your browser. Nothing is uploaded, and it keeps working with the network off once the page has loaded.

The part I spent most of the time on is the check at the end. Before it lets you download, it reopens its own output, pulls the text back out, and pixel-checks every box you drew on rendered pages. If anything you marked survived, the download is blocked and it shows you where.

That exists because I shipped the bug myself. A PDF whose content stream starts with an unbalanced cm transform leaves a non-identity CTM at the end, so the bars I appended afterwards were drawn scaled and flipped, landing nowhere near what I had marked. On a scanned page the text-based verification had nothing to catch it with, because an image page has no text, so the output looked perfectly fine.

It also flattens scanned pages and drops the invisible OCR text layer that keeps "redacted" scans searchable, and strips metadata, bookmarks and annotations.

Free for 3 documents a month, no account. $29 once for unlimited.

Built with pdf-lib and pdf.js.