Cover data in a screenshot
Paste or load the screenshot, drag a box over the data you want hidden and download the image: under the box the pixels are replaced, not covered by a layer.
PNG, JPG or WebP. The result is always a PNG, at the same pixel size as the original.
Add a box by typing its measurements
Useful if you work from the keyboard, or if you want to cover an area to the pixel. Measurements are in image pixels, counted from the top left corner.
Why a box, and not a blur
Blurring and pixelating look safe and are not, and the case where they betray you most is exactly the one they get used for most: text. Reading a pixelated caption back does not require inverting anything: you try the possible strings, pixelate them the same way with the same font and keep the one that gives the same image. With an account number or a plate, which have a fixed shape and few characters, the attempts to make are very few.
A solid box cannot be undone, because nothing is left to start from. And it has a second virtue, which is half the job: it shows. A solid bar tells whoever is looking that data was there and that it was taken away, while a blurred patch just looks like a flaw in the picture. If you want pixelation as a style and not to hide anything, then it is time to use Pixelate image.
Under the box nothing is left
When you press the button the image is not decorated: it is rebuilt. The original screenshot is redrawn at its true size onto a fresh canvas, the boxes are printed on top of it, and the PNG is written from that canvas. In the file you download the pixels underneath are not there, and there is no layer to move, to switch off or to peek under.
This is the difference that really counts, because the most common way to get it wrong is to put a box over a document and save it in a format that keeps the objects separate: there the bar can be moved and the text floats back up, and it happens every time somebody publishes a badly redacted record. If the document to cover is a PDF, this is not the right place and the job is done by Redact a PDF.
What this tool does not do
It does not decide for you what counts as data to hide: you place the boxes, and an area left uncovered stays uncovered. Before downloading it is worth looking at the whole image, because in screenshots data repeats in places nobody thinks about: the window title, a notification that arrived at the top, a chat preview in the taskbar, the file name.
It does not hide what sits outside the pixels either, and on that there is some good news. The hidden data of the original, such as the date, the phone model and the location, does not carry over to the new file, precisely because the image is rebuilt from scratch. If instead you have a photo to clean up and there is nothing to cover, the short way is Remove metadata.
Measurements typed by hand, and the keyboard
Dragging is fine with a mouse and with a finger, but with a keyboard there is no dragging: that is why a box can also be added by typing its four measurements in pixels, counted from the top left corner. It also helps when you have to cover an area to the pixel, for example the same strip on twenty screenshots in a row. Every box you place appears in the list under the picture, with its measurements and a button to remove it.