Honeytoken
A honeytoken is a planted link, credential, or record no legitimate user should ever touch. Any access to it is deterministic proof of a bot or intruder.
A honeytoken is a piece of planted data: a hidden link, a fake credential, a decoy API key, a canary record. That has no legitimate purpose. No real user can find it; no honest crawler will follow it. So any interaction with it is deterministic proof of automation or intrusion.
Why determinism matters
Most bot detection is probabilistic: signals are scored, thresholds are tuned, and there’s always a gray zone. A honeytoken hit is different in kind. It’s a fact, not a score. The client requested a URL that exists nowhere a human could see it. No fingerprint quality, no residential proxy, no browser-perfect automation changes that.
That certainty is what makes honeytokens ideal triggers for enforcement: acting on a honeytoken hit carries none of the false-positive risk that makes site owners hesitant to act on behavioral scores.
How WebDecoy uses them
WebDecoy plants honeytoken decoy links that are off-screen, aria-hidden, and nofollow: invisible to users, skipped by search engines, found only by scrapers parsing raw HTML. A hit produces a detection with ground-truth confidence, and under session clearance, it’s what revokes a bot’s access on every IP it rotates to. The same primitive is available in the open-source SDK as a tripwire you can run in your own Express app. And an SDK tripwire hit feeds the identical clearance deny path as a hosted decoy, forwarding the session’s token so the lockout is durable and rotation-proof.
Related terms
- Session Clearance: the enforcement a honeytoken hit triggers
- JA4 Fingerprint. Probabilistic identity, complementary to deterministic traps
Frequently Asked Questions
What's the difference between a honeypot and a honeytoken? +
A honeypot is a decoy system or endpoint that attracts attackers; a honeytoken is a decoy piece of data: a link, credential, API key, or record, planted inside a real system. Both work on the same principle: no legitimate activity ever touches them, so any interaction is a true positive.
Can honeytokens cause false positives? +
Properly deployed, effectively none. A well-placed honeytoken link is invisible to humans (off-screen, aria-hidden) and marked nofollow so search engines skip it. The only clients that find it are the ones parsing your page in ways no browser renders, which is exactly the population you want to catch.
See these concepts in action
WebDecoy puts deterministic detection and rotation-proof enforcement behind a 5-minute setup.