Toolpit
Loading your page…
Toolpit
Loading your page…
Find and delete a watermark, keeping the text.
No file to hand?(sample-watermarked.pdf · made-up data)
Most watermarks are a separate object in the file — a stamp, a layer, a line of text, or an image repeated on every page. Toolpit reads the document’s structure, shows you exactly what it found, and deletes only what you tick. Because it removes the object rather than painting over it, your document comes back with its text still selectable and searchable. If a mark turns out to be baked into a scan, the tool says so instead of quietly doing nothing.
Drop in the watermarked PDF — it is scanned as soon as it lands.
Check what the scan found, and tick what you want gone.
Watch the preview update, then download the clean PDF.
No. Toolpit does not upload your file for processing — the tool runs as code in this browser tab, works on the copy your browser opened, and forgets it when you close the tab. You can check this yourself: open your browser’s network panel and run the tool. You will see the page, its scripts and the ads on the site, and — when you save, if you are signed in — one small request that charges 10 tokens to your account. None of them carries your file, its name or anything read from it, because file processing stays on your device.
Painting over it makes the page a picture and takes your text with it. This deletes the watermark from the file’s structure — the stamp annotation, the layer, the text run, or the repeated image — so everything else is untouched and your document’s text is still selectable and searchable afterwards.
Because there is nothing to find. If the file came from a scanner, or was flattened or re-rendered by whoever sent it, the watermark stopped being a separate object and became part of the page image. No tool can cleanly delete it at that point; the only options are to paint over it or to leave it, and the tool tells you which case you are in rather than pretending to work.
Only if you tick it. Anything repeated on most pages gets listed — a running header repeats just as reliably as a watermark does — so findings are ranked, and only the ones that clearly read as a watermark are ticked for you. Everything else sits under “probably real content” and is left alone unless you say otherwise.
That is between you and whoever put it there. This is the ordinary case of taking a DRAFT stamp off your own document, or a mark you added yourself — a watermark is a label, not a licence, and removing one does not change who owns the underlying work.
Every one of these runs in your browser too — no upload, no watermark.