Browser fingerprint tool shows how easy you are to track using the latest sneaky tricks
A developer has released Glassbox, a free tool that runs entirely in a user's browser to reveal how easily companies can fingerprint and track individuals online, without sending any data to external servers. Built by security engineer David Dale with assistance from Claude Code, the tool matters because it exposes the many subtle techniques, from audio analysis to canvas rendering, that trackers use to identify users even without cookies, and it offers practical advice on reducing that exposure.
Glassbox checks more than 30 signals, including canvas, WebGL, font libraries, WASM features and cross-site login states, including a newly noted technique from AliExpress that uses silent audio waveforms to identify browsers. Testing by The Register found identifiability estimates ranging from 99 percent in Chrome down to 56 percent in Tor Browser, with Firefox and Tor the only browsers that avoided being labelled "effectively unique". Dale stressed the figures are estimates rather than precise measurements, since the tool has no live population data to compare against, unlike rivals AmIUnique and EFF's Cover Your Tracks. His key advice is that blending into a large crowd of similar browser configurations, rather than using a heavily customised "hardened" setup, offers the best protection, alongside using a VPN or Tor and blocking WebRTC leaks.
- New tool Glassbox reveals how trackers fingerprint your browser
- Runs locally, estimates ranged from 56% to 99% identifiable
- Advice: blend in with common setups rather than "hardening" your browser
New here? Start with this
Cookies are small files that websites can use to remember visitors, but companies can also recognise browsers by combining details about how they work. These details, called a browser fingerprint, can include graphics and audio behaviour, installed fonts and available software features.
Glassbox is a free tool made by security engineer David Dale. It checks more than 30 browser signals on the user’s device and is designed to show how distinctive that browser may appear, without sending the results to an external server.
Fingerprint estimates depend on how many other browsers can be compared, so they are not exact measures of how likely someone is to be tracked. The issue matters because a distinctive setup may be easier to recognise across visits; using common browser settings can make a person’s fingerprint less unusual.