Raw observational data
A breakdown of low-level behavioral traces: pointer jitter, keystroke cadence, correction rhythm, scroll timing, and focus changes. Each card states what is observed, how it is stored, and where the claim fails.
The first field report characterizes the session. The second describes the data supply chain.
Remote analysis unavailable. No analysis text was generated.
This is the server-side view of the same page load: request headers, IP routing, coarse MaxMind network context, and the User-Agent parser.
IP address
—
Every server you connect to logs this by default. Advertisers buy IP-linked household graphs that tie your phone, laptop, and TV together without a single cookie.
Coarse location
—
Paired with your device fingerprints, this is enough to lock a 'household' of devices to one address — even across different browsers and private windows.
Latitude / longitude
—
Precise enough for ad-tech geofencing and store-visit attribution — matching this session to a specific building, not just a city.
ISP / ASN
—
The first thing most fraud- and bot-scoring systems check: residential, corporate, VPN, or data-center traffic, each priced and treated differently.
Browser / OS / language
—
The baseline every other fingerprint gets checked against — a mismatch here is exactly what flags a spoofed or automated session.
FingerprintJS and ClientJS convert ordinary surfaces into a repeatable identifier: canvas, WebGL, audio, fonts, mime types, screen geometry, CPU threads, and memory.
The identifier is derived from the machine shape, not a login field.
Sold to data brokers and ad exchanges as a stable ID that survives cookie clearing and private browsing — the fallback tracking method once cookies get blocked.
Bot-detection and ad-fraud vendors check the exact GPU/CPU signature to catch emulators, click farms, and headless browsers pretending to be real devices.
Screen and viewport combinations narrow down the device model fast enough to help match this session to a specific make and generation, no login required.
Font and plugin lists are classic fingerprint entropy — anti-fraud systems flag a mismatch here as a sign of a spoofed or automated browser.
Canvas hashing and the webdriver flag are core anti-bot signals. Do Not Track is logged, not honored, and gets used the same way as everything else here.
Companies do not need your name to recognize the pattern. Pointer motion, scroll rhythm, typing cadence, pauses, corrections, and device shape can be turned into a behavioral fingerprint, then scored as familiar, risky, automated, or suspicious.
Fingerprint confidence
70%
Low confidence is still useful evidence: this browser limits some fingerprint surfaces.
Pointer contact, scroll movement, and typing cadence are appended with wall-clock time. The list is not interpretive. It is the record forming.
Awaiting observable behavior.
We make clear what we can see, and you can inspect it too. Give the exported data to a model you trust and ask what the Internet can infer from it: not just privacy risk, but how these signals can put you in harm's way for fraud.
Copies the current telemetry object: passive exposure, fingerprint surface, network inference, event log, taps, scrolls, keystrokes, and derived summaries.
That faint trail is your own cursor — the same passive tracking every site can do, just visible for once.