Vektor T13 - Web Assembly Browser Fingerprint

Vektor T13 Technologies • detect.expert
Developed by Dmytro Momot

Single‑file, offline WebAssembly timing micro‑benchmark that produces a fingerprint‑like feature vector and a SHA‑256 hash based on measurements across the JS↔Wasm boundary (including scripted getter/setter behavior). Includes a local enrollment gallery (localStorage), identify mode (cosine similarity + Jensen–Shannon distance), and a multilingual UI (RU/EN/ZH‑CN/VI).

What it does

• Builds a tiny Wasm module at runtime (no .wasm download).
• Runs timing micro‑tests:
  • SG0: scripted getter (Wasm global read)
  • SS1 / SS2: scripted setters (argument adaptation)
  • JS→Wasm calls and Wasm→JS callback timing
• Normalizes → quantizes → down‑samples the feature vector and computes SHA‑256.
• Lets you:
  • Enroll locally (store samples in localStorage)
  • Identify against enrolled samples
  • Export / Import the gallery as JSON

Usage

1. Save the HTML file (e.g. index.html).
2. Open it in a modern browser (Chromium/Firefox).
3. Click Collect fingerprint.
4. Optionally set a label and click Enroll locally.
5. Click Identify to compare against your local gallery.

Notes for stable measurements

• Keep the tab focused (foreground).
• Avoid heavy background CPU/GPU load.
• Compare results under similar power/thermal conditions.

Privacy & ethics

This demo runs offline and does not send any data to a server.
However, fingerprinting can be used for tracking — use responsibly and comply with applicable laws and platform policies.

License

No license is included by default — add your preferred LICENSE file if you plan to distribute.