About FootprintShield by Sifatullah Siyam

Inspiration

Most internet users believe that clearing cookies, switching to Incognito mode, or turning on a VPN makes them completely invisible. However, modern advertising and surveillance networks track users through a much stealthier method: Digital Fingerprinting.

Websites can look past cookies and read a device’s hardware and browser configuration—such as its exact CPU core count, available RAM, battery drainage velocity, graphics card drivers, and audio rendering math. Because the combination of these variables is unique to almost every machine, trackers can follow users across the web without their consent. I wanted to build a tool that defeats this privacy invasion by uniforming the user's data into a single, generic profile, turning a highly identifiable device into just an anonymous face in a crowd.


How I Built It

FootprintShield is a lightweight browser extension built entirely on the Manifest V3 architecture. The implementation targets two distinct layers of protection:

1. Network Interference (background.js)

Using the browser's dynamic rule engine, the backend intercepts network requests at the application boundary. It automatically drops connections trying to reach known tracking domains like Google Analytics and DoubleClick before their scripts can even download.

2. Deep Context Injection (content.js)

To manipulate hardware variables that are usually read-only, the extension injects an isolated script directly into the page context at document_start. This overrides default browser global objects:

  • Hardware & OS: Standardizes the reported configuration to a quad-core CPU ($4\text{ cores}$) and an $8\text{ GB}$ RAM memory limit running a specific Linux build.
  • Power Dynamics: Completely overrides the Battery Status API to read a constant state of $100\%$ capacity and charging status ($1.0\text{ level}$), preventing tracking via battery decay curves.
  • Graphical and Audio Fuzzing: Modifies data calls to Canvas 2D and WebGL rendering APIs. It introduces controlled mathematical noise into pixel data extraction and audio frequency processing to permanently corrupt the tracker's fingerprinting algorithms.

Challenges I Faced

The biggest technical hurdle was ensuring the data spoofing occurred before any tracker scripts executed on the host webpage. In Manifest V3, content scripts run alongside web scripts, which often causes a race condition. If a tracker reads the hardware variables before the extension overrides them, the defense fails. I solved this by injecting a raw string representation of our masking function directly into a dynamically generated DOM <script> element and inserting it directly at the root of the page structure, ensuring our code evaluates instantly.

Another challenge was adding noise to Canvas and Audio outputs without breaking legitimate webpage features (like video streaming or interactive elements). I had to fine-tune the injected noise to alter only a microscopic portion of the data array, making it completely invisible to human eyes and ears but entirely chaotic to tracking algorithms.

What I Learned

Through this project, I gained a deep understanding of browser security models, the DOM execution lifecycle, and JavaScript injection techniques. I learned that data privacy in 2026 isn't just about blocking tracking cookies. it requires actively managing the behavioral and hardware properties that your machine broadcasts to the outside world. This project proved that creative engineering can reclaim user privacy even against highly sophisticated automated corporate tracking networks.

Built With

Share this project:

Updates

Submission history