Drop a screenshot or photo above, pick blur, pixelate, or blackout, then drag rectangles over anything you don't want the world to see — faces, API keys, auth tokens, email addresses, account numbers, home addresses in map views. Stack as many regions as you need, mix modes per region, and delete or undo any of them before downloading the redacted copy at full resolution. Everything runs on your device — the image is never uploaded to a server, which matters when the whole point is that it contains something sensitive.
One honest caveat: blur and pixelation at low strength can sometimes be reversed. Researchers have recovered text from lightly pixelated screenshots by comparing mosaic patterns against rendered candidates, and mild blurs leave enough signal for deconvolution. Crank the strength slider up for anything that matters, and for truly sensitive text — passwords, keys, secrets — use blackout, which replaces the pixels with solid black and leaves nothing to recover. Our post on why pixelation is not redaction walks through how these attacks work.
Redaction pairs well with a couple of neighbouring cleanups: the image cropper cuts away whole areas you don't need to share at all, the EXIF remover strips hidden GPS coordinates and camera metadata from photos, and the image compressor shrinks the final file before you post it. Redacting a document instead of an image? The PDF redactor blacks out content so it is truly removed from the file.
Frequently asked questions
Can pixelated or blurred text be reversed?
Sometimes, yes — especially text. Research tools like Depix and Unredacter have recovered passwords from pixelated screenshots by rendering candidate characters, pixelating them the same way, and matching the mosaic patterns; light blurs leave enough signal for deconvolution attacks too. If the region matters, crank the strength slider up, and for genuinely sensitive text use blackout instead — it replaces the pixels entirely.
What should I use to hide API keys, passwords, or SSNs?
Blackout. It overwrites the region with solid black pixels, so there is literally no information left in the image to recover — unlike blur and pixelation, which transform the original pixels but keep a residue of them. Save blur and pixelate for things where looks matter more than secrecy, like faces of bystanders or UI clutter.
Why do most tools default to blur instead of blackout?
Aesthetics. A blurred or pixelated region looks softer and keeps the visual context of the screenshot, while a hard black rectangle looks severe — so blur became the screenshot-sharing convention. That convention is about how the image looks, not how secure it is, which is why this tool offers all three modes and lets you mix them per region.
Is my image saved anywhere?
No. The image is decoded and edited entirely in your browser using a canvas — it is never uploaded to a server, and nothing is stored after you close the tab. The only output is the redacted file you download, which matters when the whole point is that the image contains something sensitive.