Open a screenshot to start redacting

Drag a file onto this window, paste with ⌘/Ctrl V, or pick one below. Decoding and editing happen inside this browser tab.

PNG · JPEG · WebP · GIF | up to 40 megapixels

Tool options Solid fill

Fill

Amount

Solid fill is the only mode no de-blurring trick can reverse. Use it on addresses, recovery words and figures.

Areas 0

Drag across the image to add one.

Output

The export is redrawn from scratch, so the original file's camera, GPS and app metadata does not come along.

Offline-capable editor No image Tool: Solid fill We will never request your recovery phrase

Before-you-post checklist

Most doxxing starts with something sitting in the corner of an image. Scan for these:

Wallet addresses. A fragment next to a balance is often enough to find you on-chain. Cover the whole string.
Balances and P&L. Large figures invite phishing DMs and worse. Hide totals, token amounts and the fiat line.
Signatures and timestamps. One transaction ID or an exact time leads straight to your account in any explorer.
Names, emails, avatars. Exchange display names, a logged-in email, profile pictures, notification banners.
Tabs and bookmarks. Tab titles and the bookmark bar reveal which exchanges and dashboards you use.
Recovery phrases. Do not photograph them at all, blurred or not. No genuine support agent will ever ask for one, and neither will we.

Questions & answers

Does my image get uploaded?

No. It is decoded onto a canvas in this tab. The Content-Security-Policy only allows connections back to this page's own origin, and there is no upload route on the server. Close the tab and it is gone.

Solid fill, mosaic or soften: which one?

Solid fill overwrites the pixels completely. Mosaic and soften leave a smeared version behind, and for short strings like addresses or amounts, reconstruction tools can sometimes guess the original. Use those two for faces and backgrounds only.

Is the original metadata kept?

No. The PNG is painted fresh from the canvas, so EXIF, GPS and editor tags in the source file are not copied over.

Can it detect addresses automatically?

Not at the moment. Text detection would mean shipping a multi-megabyte OCR engine to your browser. Marking areas by hand is also the surest way to know nothing slipped through.