Privacy Checker

If EXIF holds a latitude and longitude, they are surfaced and a map link to those coordinates is generated. Being told "a GPS tag is present" does not tell you how bad it is; opening the map and seeing your own home does. Altitude and the north/south, east/west reference values are checked alongside the coordinates.

Whether a photo is geotagged depends on the camera app's settings and on the route the file took to reach you. The point of this check is to catch the cases where "I'm sure location is turned off on my phone" turns out to be wrong.

Capture time, digitisation time, and modification time are checked, along with the date and time stamps that come from GPS. A timestamp looks harmless on its own; line several posts up and it describes a routine — which hours you are reliably not at home.

Device and camera model, lens, and exposure settings are shown too. These can be enough to link images you believed were posted anonymously. The full EXIF record can be copied out as JSON if you want to keep it.

This is the check people overlook most and the one that costs the most. Cameras commonly embed a reduced-size copy of the image inside EXIF at capture time. The catch is that when you later crop the picture or paint over part of it, some editors never regenerate that embedded thumbnail.

When that happens, the area you cropped away — or the region you thought you had covered — survives inside the thumbnail in its original state, and anyone who receives the file can pull the pre-edit composition back out. This tool does not merely report that a thumbnail exists: it renders it for you to look at, so you can compare it against the visible image and judge whether they disagree.

For what it is worth, OFFiler's own image tools — compression, resizing, format conversion, EXIF removal — redraw through a canvas, which discards EXIF and the thumbnail along with it. Images that have been through those tools do not carry this risk.

Any QR code or barcode caught in the frame is detected, and the string encoded inside it is shown to you. A delivery label on the desk behind you, an employee badge, an event ticket, a Wi-Fi join code — things that read as abstract patterns to a human eye but decode into personal or credential data.

It is not limited to QR codes: Code 128, Code 39, Code 93, EAN-13, EAN-8, UPC-A, UPC-E, ITF, PDF417, Data Matrix, and Aztec are all covered — with a browser caveat explained next.

Code detection first tries the barcode detection facility built into the browser itself. Chrome, Edge, and Android browsers ship this, and it reads every format listed above quickly.

Where that is unavailable — Safari and Firefox — the tool falls back automatically to a QR-parsing library bundled with the site. That fallback handles QR codes only; other barcode families are not detected. The same image can therefore produce different results in Chrome and in Safari. If you need certainty about barcodes, run the check in a Chromium-based browser. Most tools leave this difference unstated; it is stated here deliberately.

Parsing EXIF, pulling out the thumbnail, and scanning for codes all happen by reading the selected image in your device's memory. Neither the file nor the coordinates and strings recovered from it are transmitted. The map link only travels to a mapping service if you click it; leave it alone and the coordinates stay put.

Besides drag and drop, you can paste an image straight from the clipboard with Ctrl+V — take a screenshot and inspect it immediately.

What is examined is the data inside the file and the codes a machine can read. A nameplate in the background, the view through a window, text on a document on the desk, a logo on a uniform, a reflection in someone's eye — routes to identifying a place or a person through what is *depicted* are outside its reach. That judgement still needs your eyes.

When you do spot something, black it out or blur it with image markup. Those marks are painted onto the pixels, so once the image is saved the covered information cannot be recovered.

Some platforms strip EXIF when you upload. But that behaviour differs between services, changes over time, and does not apply at all to email attachments, files sent directly in chat, cloud-storage shares, or marketplace listing photos.

Relying on "I hear that service handles it" is a fragile assumption. The reliable version is to hand over a file that contains nothing in the first place: if the check turns something up, run it through EXIF removal before publishing.

Route the fix by what came back: EXIF removal for location and timestamps, markup for things caught in frame, both when both apply. Compressing or resizing for publication drops EXIF as a side effect anyway.

The reason to check *first* is to learn what the problem was. Compress straight away and the EXIF goes, but you never find out that the photo was geotagged — and therefore never realise that every other photo from that device is in the same state.