Redact a PDF, and see it proved

Mark what has to go. Every page is rebuilt as an image, so the words are destroyed rather than covered — and the finished file is read back and searched before you are offered a download.

On this device

Mark what has to disappear

Search for a name or number, or drag across the page. Nothing is uploaded: the document is opened by this page, on this device.

    Opened on this device · never transmitted

    What this tool does, exactly

    What it does
    Draws every page of your PDF as an image, paints your marks onto those pixels, and builds a new document from the images. Text, vector drawings, annotations, form fields, attachments and metadata are not carried over, because nothing is carried over — the file is written from scratch.
    What it proves
    Before you are offered anything, the finished file is reopened from its own bytes and searched. Five measurements have to come back clean; you see all five, passed or failed, with the number each one produced.
    What it cannot know
    Which words matter. It removes what you mark and nothing else, so read the result before you send it — a name you did not mark is still in the document.
    What it costs you
    The text layer. The finished file is larger and its words can no longer be selected, searched or read aloud. Keep your original; this is the copy to send, not the copy to archive.

    About this tool

    Is a black box drawn over the text really a redaction?

    No, and that is the mistake this tool exists to avoid. A rectangle drawn on top is a separate object; the words underneath stay in the file and can be selected, searched and copied out. This tool converts each page to an image and rebuilds the document from those images, so the text is gone rather than hidden.

    What is the check that runs before the download?

    The finished file is opened again, as a stranger receiving it would open it. Its text is extracted and every phrase you marked is searched for. The download appears only if there are zero matches, no extractable text at all, no annotations, form fields or attachments left, and every mark is solid black in the finished page image.

    What happens if the check fails?

    You get no file. The result panel says which check failed and why, and nothing is offered for download, because a redaction that cannot be proved is worse than none — it is one you would have sent.

    Does my document leave my device?

    No. There is no server to send it to: this site is static files, and the work runs in your browser. You can confirm it by opening your browser's network panel, or by disconnecting from the internet after the page has loaded — the tool keeps working.

    Why is the finished file bigger, and why can I no longer select the text?

    Because it is now made of page images. That is the trade: a redaction you can rely on costs you the text layer and some file size. If you need selectable text elsewhere in the document, keep your original — this file is the one that is safe to send.

    Does this remove metadata and hidden data too?

    Yes. The new document is built from scratch, so document metadata, annotations, form fields, attachments and any document-level scripts are not carried over. The finished file has no title, author, producer or timestamps.

    More secure tools · All 10 · How they work