b2KIT

Canary Token Generator

Generate honeypot tokens (URLs, DNS names, email addresses, files) that alert you when accessed by an attacker.

Tested tool guide Tested browser tools Checked August 16, 2026

What Canary Token Generator does and how it behaves

This tool creates canary tokens: unique markers you hide in places an attacker might touch. Choose a carrier, a URL, a DNS name, an email address, or a file, and the tool generates a token with its own identifier, entirely in your browser, nothing about your setup is uploaded. When someone later fetches that URL, resolves that name, mails that address, or opens that document, you receive an alert. The surprise most people hit: generating the token is the easy part. It does nothing until deployed where an attacker can find it, and the alert only reaches you if its notification channel is configured.

How the result is produced

1

Unique identifier per token

At creation, each token receives an identifier different from every other token. The tool formats that identifier into the carrier you selected: a URL path, a DNS name, an email address, or a file. The carrier looks like the ordinary thing it pretends to be, so it attracts no special attention, and the embedded identifier is what makes a later hit traceable to this one token.

2

Alert fires only on contact

A token alerts when something actually touches it: an HTTP request to the URL, a DNS lookup of the name, mail delivered to the address, or a document that fetches a remote resource. The alert is then delivered through the notification channel attached to the token. A token that is never deployed, or deployed where no listener can observe it, stays silent.

Good uses

  • Track a document you send out: a contract or spec that contains a token notifies you the moment it is opened, forwarded, or leaked, even though it never comes back through you.
  • Decoy material for intrusion response: a fake credentials file or database dump planted where an intruder would grab it flags the exfiltration while it is still happening.
  • Map network reconnaissance and harvesting: a DNS name or email address token embedded on a public page tells you when your infrastructure is being enumerated or your addresses collected for spam and phishing campaigns.

Limits and checks

  • A tripped token is not proof of an attack. Link previewers, antivirus scanners, web crawlers, and curious colleagues can all fire a token. Treat an alert as a reason to investigate, not as a confirmed compromise.
  • Silence is ambiguous. Sandboxed mail clients and many PDF or Office viewers refuse to load remote content, so a document token can be read without ever alerting. You cannot distinguish 'never touched' from 'touched but the request was blocked'.
  • An alert identifies the token, not the person: the access may have been automated, proxied, or made from an unrelated location, and if you embed the same token in several copies you will not know which one tripped. Keep a log of where each copy goes.

Common questions

I opened the document myself and got no alert. Is the token broken?

Probably not, but verify before relying on it. Many readers never fetch remote content: email preview panes, PDF viewers, and sandboxed Office apps commonly block it, so nothing triggered the token. It can also mean the notification channel is misconfigured or filtered as spam. Trigger the token deliberately with a normal browser, and confirm the alert arrives, before you trust the setup.

Can an attacker spot the token and remove it?

Yes, sometimes. Tokens are designed to look unremarkable, but someone who inspects a document's internals or a URL's hostname can recognize a marker that points to a monitoring service and strip it out. That is not a failure of the design: a canary's value is the alert you get when a file is touched that you expected nobody to touch, not secrecy.

References and verification

The behavioral notes were checked against the browser implementation. Standards and primary references below define the relevant format, formula, or platform behavior.

Related Tools