See what a screenshot still reveals before you post it.
Or just paste one — take a shot and press ⌘V. Nothing is uploaded.
Scanning the image…
Nothing was recognised by the checks below. That is not the same as the screenshot being safe to post — it is the answer to the questions this tool knows how to ask.
Your screenshot is decoded, analysed and redacted in this browser tab. It is never uploaded.
Before a screenshot goes into a message, a document or a public post, it is worth knowing what somebody at the other end could still read in it. Drop one here and this lists everything it can recognise: email addresses, account names, URL bars with tokens in them, notification badges with numbers on them, and anything else that looks like it was not meant for public view.
The things that give people away are always the same. A tab bar with an inbox count. A sidebar naming Slack channels and team members. A URL with a session token or an internal hostname. The clock and status bar that names the device. None of these are secret in isolation, and all of them together narrow down who took the screenshot.
Everything here runs on your device. The image is never uploaded, and when you black something out the original pixels are overwritten in the canvas before the file is saved. There is no version of this that says your screenshot is clean — it tells you what it checked, what it found, and what it could not look at, because no list of things to look for is ever finished.
Support threads, bug reports, product posts — all of them ask for screenshots of things only you can see. The URL bar, the account menu and the notification count are all in there.
A funny message, a support reply, an egregious email. The sender's address, the timestamp and the surrounding messages are still in the frame.
Product documentation and changelog screenshots often carry internal URLs, staging hostnames, test account names and developer toolbars nobody noticed.
Whistleblowers and journalists share screenshots of internal tools. The status bar, the time zone, the account selector and the notification badges all point back at the source.
Take a screenshot and press ⌘V, or drag the file in. The image is read in this browser tab and nothing is uploaded.
Each row is something recognisable in the image: an email address, a name, a URL, a notification badge. Underneath, what the tool checked for and what it could not look at.
One click blacks it out. You can also draw your own boxes over anything the tool did not recognise. The redaction overwrites the pixels — it is not a layer that can be removed.
A PNG with the original pixels gone. If you want to make it look presentable afterwards, the next step is Screenshot Beautifier.
No. The image is decoded and searched in this browser tab, and the redacted copy is built here too. Turn your wifi off after the page loads and it still works — which is worth doing once, given what people share.
Email addresses, URLs with query strings or tokens, account names that look like identifiers, notification badges with numbers, status bars and clocks. The checks are listed under the findings so you can see exactly what it covered.
It means every check listed on the page came back empty, and the page names them so you can see which ones they were. It does not mean the image is clean, because no tool can promise that — it checked the things it knows how to check.
It looks for patterns that match identifiers — email addresses, URLs, account IDs — rather than reading every word. For full text extraction, use Image to Text; this tool finds the things that might be worth hiding.
Yes. The redaction is drawn directly into the image data before the file is written. It is not a layer on top, and there is no metadata to undo it from. The pixels you black out are gone.
Yes — both options are there. A blur is enough to stop casual reading; a black box is proof that nothing remains. Pick whichever one fits what you are doing.