Explore our new Cybersecurity Terms Glossary, explained by our experts.

View the Glossary View the Glossary
PRIVIA

Cybersecurity Terms Glossary

What Is Tabnabbing?

Tabnabbing is a phishing technique in which a page the victim has left open in a background tab quietly rewrites itself into a convincing login screen while their attention is elsewhere.

Illustration for the cybersecurity term Tabnabbing
Home / Blog / Cybersecurity Terms / What Is Tabnabbing?
17 November 2026

What Is Tabnabbing?

What Is Tabnabbing?

Tabnabbing takes advantage of a habit almost everyone has: leaving a dozen tabs open and not looking at most of them for hours. The attack lets a page that was harmless when you opened it change into a login screen for a service you trust, so that when you return to the tab you assume your session expired and type your password into the attacker’s form.

Aza Raskin, then at Mozilla, described the technique in May 2010. It is a user interface attack rather than a software vulnerability: nothing in the browser malfunctions, and the deception relies entirely on the fact that people identify a tab by its favicon and title rather than by re-reading the address bar.

The Classic Version

The sequence is short:

  1. The victim opens an attacker-controlled page, perhaps from a forum link or a search result. The page looks ordinary.
  2. The victim switches to another tab. The page detects this through the Page Visibility API, a blur event, or simply a timer.
  3. After a delay, the page replaces its own content, title, and favicon with a replica of a well-known sign-in screen, such as a mail provider or a corporate portal.
  4. The victim returns, recognises the branding, assumes they were logged out, and submits their credentials.
  5. The page stores the credentials and redirects to the genuine site, which usually shows the victim as still signed in, so nothing feels wrong.

The only reliable tell is the address bar, which still shows the attacker’s domain. That is precisely the element users stop checking once a tab has been open for a while.

Reverse Tabnabbing

The more damaging variant works in the opposite direction and does not need the victim to visit a malicious site at all. When a page opens a link with target="_blank", the new page historically received a window.opener reference to the page that opened it. Cross-origin script access is blocked, but one property was not: the new page could set window.opener.location and navigate the original tab anywhere it liked.

So a user on a trusted site, such as a forum, a webmail client, or an intranet portal, clicks an external link. The external page sends their original tab to a convincing copy of that trusted site. The user finishes with the new tab, closes it, finds the old tab apparently logged out, and signs in again.

Browsers fixed the default in 2020 and 2021. Safari 12.1, Firefox 79, and Chrome 88 made target="_blank" imply rel="noopener", so window.opener is null. The problem has not disappeared, because a great deal of software renders links in older components, in email clients, in desktop applications embedding a web view, and in content authored with an explicit rel="opener".

Defences for Developers

OWASP lists reverse tabnabbing in its guidance on HTML5 security, and the fix appears in most secure coding checklists as a one-line change that is frequently missed in older codebases.

Defences for Users and Organisations

Credential theft is the goal, so the controls that blunt phishing generally apply. Password managers are particularly effective here: they fill credentials only on a matching origin, so a replica page on a different domain simply does not autofill, which is a far more reliable signal than human vigilance. Phishing-resistant multi-factor authentication, such as FIDO2 security keys or passkeys, binds the authentication to the real origin and makes a captured password insufficient.

In security awareness training, the useful habit to teach is narrow and memorable: if a tab you did not just open is asking you to sign in, close it and navigate to the service yourself. The NCSC takes the same line in its phishing guidance, placing the emphasis on technical controls and easy reporting rather than on expecting staff to spot every fake.

At Privia Security, our Social Engineering Services include browser-based credential harvesting scenarios like this one, so you can see how your staff and your controls respond before a real attacker tries.

FAQ – Frequently Asked Questions

Is Tabnabbing Still Possible in Modern Browsers?

Classic tabnabbing is, because nothing stops a page from changing its own content, title, and favicon after you switch away. That behaviour is normal web functionality. Reverse tabnabbing is largely blocked in current browsers, which treat target=”_blank” as implying rel=”noopener”. Older browsers, embedded web views, and desktop applications that render HTML can still be affected.

What Does rel=”noopener noreferrer” Actually Do?

noopener makes window.opener null in the newly opened page, so it cannot navigate or inspect the page that opened it. noreferrer does the same and additionally suppresses the Referer header, so the destination does not learn where the user came from. Using both together is the standard recommendation for any link that opens in a new tab.

How Does a Password Manager Help Against Tabnabbing?

A password manager matches saved credentials to the origin in the address bar, not to the appearance of the page. A replica login screen hosted on a different domain will not trigger autofill. That silence is a strong signal, and it works even when the user has not noticed anything unusual, which is why managers outperform visual inspection.

Can a Content Security Policy Prevent Tabnabbing?

Not directly, because the attacking page controls its own policy. The relevant header is Cross-Origin-Opener-Policy, which when set to same-origin isolates your pages from any window that opened them or that they open. Combine it with rel=”noopener” on outbound links so that users on browsers without full COOP support are still protected.

Related Terms

You May Be Interested In These

What Is Log Management?

Cybersecurity Terms

17 November 2026

What Is Log Management?

Read More
What Is LLM Security?

Cybersecurity Terms

17 November 2026

What Is LLM Security?

Read More