Before you send a file anywhere, it is worth knowing what travels with it: the coordinates where it was taken, the device, the editor, sometimes the prompt and a signed provenance record. This viewer reads those fields on your device and lists them with a risk level. It does not modify the file and it does not upload it — reading first is the whole point.
Engine
File API
Runs in
Web Worker
Needs account
No
Files differ. A photo straight off a phone carries a dozen fields; the same photo after a trip through a chat app usually carries none. The table below is the typical inventory, not a promise for any single file.
Editors, researchers and platform moderators receive files from strangers. The metadata is the only part of the file that makes a claim about its own origin — who shot it, on what device, when, and whether a tool signed it. Reading it before publishing is basic hygiene.
A listing photo taken in your flat often carries the coordinates of your flat. Buyers do not need them and scammers find them useful. Look first, then decide whether the file you upload should have them at all.
Instagram, WhatsApp and most marketplaces remove metadata on upload. Knowing that your file no longer carries GPS is different from assuming it. The viewer tells you what is actually in the copy you hold.
A C2PA manifest, an IPTC AI label and an EXIF software field are all claims written by whoever saved the file. Reading them tells you what was claimed. It does not tell you whether the claim is true, and this page will not pretend otherwise.
Drop a photo or video into the panel. It is parsed in a Web Worker on your device — the network panel stays quiet.
Read the list. High-risk fields are the ones that identify a person or a place; the raw value is shown so you can judge it yourself.
If you want them gone, open the matching remover and download a cleaned copy. Viewing here never changes the original.
Before you rely on it
Most image generators never write a C2PA manifest, and every social platform re-encodes uploads. A file with an empty report is the normal case, not a verdict. Treat “nothing found” as “this copy carries no claims”, and nothing more.
It is the same reading engine. The remover runs the same inspection and then rewrites the container without the metadata; this page stops after the report so you can look without committing to a change.
No. The file is read with the browser’s File API and parsed in a Web Worker. There is no upload endpoint in the site, and the viewer keeps working with the network disabled after the page loads.
The app that last saved it probably removed the metadata. Messaging apps, social platforms and screenshot tools all do this. It is also possible the camera never wrote any — many front-facing modes skip GPS.
No. It reads claims: an IPTC digital source type, a generator tag, a C2PA manifest, a prompt left in a PNG chunk. A file can be AI-generated with none of those, and a real photo can carry a label that says otherwise.
Yes — the viewer reads the container structure and metadata boxes, not the video samples, so a 400 MB clip is inspected without decoding a single frame. The file size limit exists only because the whole file must be held in memory.