Free · no sign-up · under a minute
Browser fingerprint test
See which of your browser's signals agree, which contradict each other, and what a detector would conclude from it.
- Checks
- 295 checks
- Subsystems
- 16 browser subsystems
- Reference
- 1 page per check
Audit this browser's fingerprint
Runs the full suite, summarizes what needs attention, and keeps every raw measurement available for inspection.
Skip the steps below and 18 input checks are reported as not attempted, never as passed. Do them first to include them.
Optional, before you run: measure how this browser handles input
0 of 5 done · adds 18 checks
- Click the three markers
- Type
audit - Drag the slider past 800
- Press and hold for 1.2s
- Your input device
Why these five, and what is recorded›
Clicks. A click you make is trusted; one a driver injects around the input pipeline is not.
Typing. A keypress is the default action of a keydown. Some drivers send only the character, leaving a keypress with nothing behind it.
The slider. A drag is the one interaction that produces a stream of values rather than a single jump, and the only thing here that asks the pointer to travel while a button is held.
The hold. A click is over in milliseconds; a hold has to stay a press for over a second, and every pointer move inside it must still report the button as down.
Your device. Genuinely useful context — a trackpad and a mouse produce different movement, and touch produces none of it. Any answer works.
Recorded: counters and HTML tag names. No key, no code, no character, and not the text you typed — the word above is a constant, so it says nothing about you.
What this bot detection test checks
Browse all 295 checksBot detection rarely relies on one value. It checks whether a browser's signals agree: whether the user agent matches the Client Hints and the engine, whether the GPU name matches what the canvas draws, whether a worker reports the same values as the page, and whether clicks and keys came through the browser's own input pipeline. This test runs that kind of comparison in your browser, 295 checks across 16 subsystems. Each has its own page in the check reference; for background, read what bot detection is and what browser fingerprinting is.
- Render & GPU54
- Automation surface51
- Navigator identity39
- Environment & locale28
- Screen & display20
- Capability surfaces20
- Native function integrity15
- Engine & OS oracles14
- Network & WebRTC11
- TLS & the network layer10
- Error subsystem8
- Input & behaviour7
- HTTP request vs the runtime6
- Clocks & ledgers6
- Worker realms4
- Connection & locale2
Understand your result
The audit stays focused on your browser. Its long-form documentation lives in dedicated, linkable references.
Fingerprint check reference
Every check, grouped by browser subsystem, with the expected result and resolution.
Read referenceMethodology, scoring & limits
How coherence is measured, which evidence is scored, and what a public page cannot prove.
Read referenceDetection research & sources
The public detector artifacts and research behind the audit's signal attributions.
Read referenceBrowser fingerprint statistics
An anonymous census of the browsers run through the test: what is common, what is rare.
Read referenceFAQ
What is a bot detection test?
A page that reads what your browser reveals and judges, the way anti-bot systems do, whether it looks automated or inconsistent. This one reports where your browser's signals contradict each other, and explains for each contradiction what a detector concludes from it and how to fix it, instead of giving a single human-or-bot verdict.
Is my browser a bot if a check fails?
No. Privacy extensions, virtual machines, unusual settings and some browsers fail checks too. A failure means two signals disagree, which a detector may weigh; each check's page says how much, and why.
How is this different from CreepJS or BrowserScan?
Those sites measure your fingerprint and flag lies and headless traits. This test compares independent parts of the browser with each other, groups the results by subsystem and explains each one. The fingerprint test guides compare what each public test site measures.
Can I run it in an automated browser?
Yes. Open this page in the browser your Playwright, Puppeteer or Selenium script drives and run the test there; it is meant for exactly that. Results are read the same way as in a normal browser.
What happens to the data?
Running the test saves your fingerprint for internal research. It is detailed enough to identify you, so it is not anonymous; it is never published, sold or shared, and you can erase it at any time. See the data notice at the bottom of the page.
Blend in. Stay clear.
A Chromium where every signal this test reads is set inside the engine, so they agree. Start with one browser, free. Keep your scripts.
pip install clearcote · npm i clearcote · dotnet add package Clearcote
Data notice: Running this test saves your fingerprint to our database for internal research. It is detailed enough to identify you, so it is not anonymous. We never publish, sell or share it, and you can erase it at any time. How this data is used