How does a safe browser extension turn malicious?
A browser extension turns malicious when a criminal buys it or phishes its developer. They push a silent update that reuses the access users already granted, so Chrome shows no new prompt.

Malicious browser extensions
A browser extension, also called an add-on, is a small program that adds a feature to your browser. Many extensions can run their own code inside every page you open, even your bank. Every extension ships with a file called manifest.json that lists what it wants to access.
Chrome reads that list and turns the risky entries into install warnings, so you can see what you are about to grant. It also keeps extensions up to date on its own, checking for new versions every few hours. Both features help users. But the update channel is also how a safe extension turns malicious.
What a browser extension actually sees
Access to websites is declared as host patterns, in host_permissions or in content_scripts.matches, and those trigger warnings. The pattern <all_urls> becomes “Read and change all your data on all websites.” The tabs permission shows “Read your browsing history.” Other entries, such as storage and cookies, show no warning at all.
That doesn’t mean cookies silently opens every cookie you have. The cookies API only works on sites the extension also holds host permissions for. The real power sits in the site access. That is the part that shows the big warning at install, and it stays granted after that.
How a trusted extension turns malicious
- It earns trust first. The add-on works honestly for months, collecting users and reviews.
- A criminal takes control. They buy the extension or phish its developer, and get the right to publish updates.
- The update installs silently. Chrome checks for a new version every few hours and installs it without asking.
- Old permissions mean no prompt. Chrome only asks again when an update adds a new warning, and keeps the extension disabled until you accept. Malicious code that reuses the old permissions gets no prompt at all.
That is what happened in December 2024. A Cyberhaven employee got a fake Chrome Web Store policy email with a “Go To Policy” button. Already signed in, they clicked “Allow” for the attackers’ OAuth (Open Authorization) app, called “Privacy Policy Extension”. That needed no password or MFA (multi-factor authentication) prompt, and it gave the app a token to publish updates. The same trick hit at least 35 extensions, about 2.6 million users.
A store takedown is not an uninstall
Many people think a bad extension disappears once the Chrome Web Store removes it. It doesn’t. Installed copies aren’t automatically removed. Chrome disables an extension it flags as malware. For other takedowns it only warns you, and the extension keeps running.
What to do
- Open the three-dot menu, then Extensions, then Manage extensions, and remove every add-on you don’t use.
- On the Details page of each remaining extension, set site access to On click.
- If an update asks for a new permission, stop and think before you accept it.
- Don’t wait for a store takedown. Check your list yourself, because an extension is only as safe as its next update.
Watch it in 60 seconds
