Hi everyone!

I wanted to share a small project I’ve been working on lately. The premise is simple: every time we share a photo or a document, we inadvertently leak a massive amount of personal data — from home GPS coordinates to camera serial numbers and even the edit history of a PDF.

Using "online privacy services" to clean your files always felt like a paradox to me (sending private data to a server to make it private?). So, I built my own tool that runs strictly locally. I call it DMS (Deceptive Metadata Shredder).

What it does
Beyond the standard "wipe everything" approach, I added a Spoofing Mode. Sometimes, having zero metadata looks suspicious or breaks the functionality of certain apps. Instead of just deleting, DMS replaces sensitive info with plausible "noise":

GPS: Injects random coordinates, but keeps them within the same country (so your photo doesn't suddenly appear to be taken in the middle of the ocean or Antarctica).

Hardware: You can pretend the photo was taken with a different camera or phone model.

Timestamps: Shifts the creation date/time if you don't want to reveal the exact moment a file was generated.

Under the hood
The project is built using Python 3.11 and leverages the power of ExifTool.
I implemented two ways to interact with it:

GUI (PySide6): Features a side-by-side "Before/After" comparison, drag-and-drop support, and a clean interface for desktop users.

CLI: For terminal enthusiasts or anyone looking to automate the process via scripts.

One of my favorite features is the "Watch Folder". You point the app to a specific directory, and any file you drop in there is automatically detected, cleaned (or spoofed), and moved to a "clean" folder. It’s a huge time-saver for batch processing.

The Battle with Antiviruses
The development process had its hurdles, specifically with Windows. Projects compiled with PyInstaller often trigger false positives in antivirus software. I had to spend some time "wizarding" with the packaging process to make VirusTotal turn green. Currently, it’s clean (save for a couple of false detections from obscure engines), so it’s safe to run on Windows without constant alerts.

Open Source
The project is completely open-source under the MIT License. If you're interested in privacy tools, Python, or just want to audit the code, feel free to check it out.

Source Code: github.com/davvikq/deceptive-metadata-shredder

I’d love to get some feedback! If you have ideas on which file formats I should add next (currently supports common images, PDFs, and Office docs), let me know in the comments.