Point a camera at the code or drop in a screenshot. Every code in the picture is decoded on this device, and links are shown in full before anything opens.
| # | Time | Source | Format | Content |
|---|
Camera frames and images are decoded in this tab and never leave your device. Links open only when you tap them.
Phone cameras read QR codes well, but only codes in front of the lens. A code inside a screenshot, a PDF you are viewing on the same phone, an email attachment or a photo taken last week is out of reach. Drop that image here and every code in it is found and labelled on the picture, including several side by side. On a laptop, the webcam works the same way, which helps when the code is on a printed page and the result is needed on the computer.
Fraudsters stick their own codes over real ones on parking meters, restaurant tables and posters, and send codes in emails because a picture of a link slips past filters that read text. This scanner never opens anything by itself. It shows the domain on its own line, warns about unencrypted links, look-alike Punycode domains and links that hide the real host behind a user name, and treats javascript: or data: content as plain text. If the domain is not the one you expect, do not open it.
The decoder is zxing-cpp compiled to WebAssembly. Besides QR codes, including Micro QR and rMQR, it reads Data Matrix, Aztec, PDF417 and MaxiCode, plus one-dimensional barcodes such as EAN-13, UPC-A, Code 128, Code 39 and ITF. It also tries rotated, mirrored and inverted (light-on-dark) codes.
A laptop cannot join a network by scanning a code the way a phone can. Take a photo of the Wi-Fi code with your phone, send it to the laptop, drop it here, then copy the password. The network name, security type and password are read from the standard WIFI: text, with escaped characters restored, so what you copy is exactly what the router expects.
Stock checks, event check-in and asset audits mean reading dozens of codes in a row. Tick Continuous scanning and the camera stays on: each new code is added to the session log with the time, and the same code is not recorded twice while it stays in view. For codes already captured in photos, select or drop a whole folder of images at once — every image is decoded in turn and each result is labelled with its file name. Export the log as a CSV that opens correctly in Excel, or copy every decoded value in one go. The log lives only in this tab and disappears when you close it.
Take a screenshot, then open this page on the same phone and choose the screenshot. The code is read from the image, so no second device is needed.
Yes. One image can contain several codes, each outlined and numbered, and you can select many images together. For a stack of physical labels, use continuous scanning and export the log.
No. Frames and images are decoded by WebAssembly inside this tab. Nothing is sent to a server and the log is not saved after you leave.
The most common causes are blur, the code being cut off at an edge, or the white margin around it missing. Try a sharper photo from slightly further away, or crop the image closer to the code.
Read the domain shown on its own line. If it is not the organisation you expect, if it is marked as look-alike (Punycode) characters, or if it hides behind a user name, type the organisation's address yourself instead.
People look for a scanner in the browser when the phone camera cannot help: the code is in a screenshot or an email on the same phone, it has to be read on a laptop, or a photo contains several codes and the camera app only ever picks one. The other reason is caution — wanting to see exactly where a code leads, in full, before anything opens.
Boarding passes, event tickets, payment requests and Wi-Fi codes do not stop working because they are inside a picture. Airline boarding-pass barcodes, for example, carry your name and booking reference. Before posting or forwarding a screenshot, decode it here to see what it gives away, then blur the code if it should stay private.
Phone cameras cannot focus closer than a certain distance, and a code that fills the whole frame often loses its quiet margin at the edges. Hold the camera a little further back so the code sits inside the frame with white space around it, and let the decoder work with fewer but sharper pixels.
A code on an official-looking sign, with a familiar logo in the middle, can point anywhere. Read the domain line in the result. A payment or parking page on a domain you have never seen, or a shortener hiding the destination, is reason to type the organisation's address yourself instead.