Skip to main content

Browser Extension

Browser Extension

Scan the page you're actually looking at, without pasting a URL. The extension runs the same engine as the web app, connected to your own account with an API key.

Manifest V3
Chrome + Firefox
Checks, same engine
905+
Page content ever collected
0

Overview

The extension is a thin client over the same scan engine and the same account as the web app: it doesn't store findings locally beyond a small recent-history cache, and it doesn't run its own copy of the checks. Two things it can do that the web app can't from a URL bar alone: react to the page you're currently on (the on-page reputation card, auto-scan on navigation) and read a page the way you're actually seeing it, cookies and all, when you ask for that explicitly.

Install

Chrome / Edge

Live

One click from the store. Updates itself from here on, no manual re-download.

Add to Chrome

Firefox

Live

One click from Firefox Add-ons. Updates itself from here on, no manual re-download.

Add to Firefox
  1. Connect your account

    Click the toolbar icon, then Open Settings. Generate an API key from your VulnRadar profile and paste it in - the extension authenticates as you from then on.

Need an API key first? Generate one from /profile?tab=developer&dtab=api-keys while logged in.

Scanning from the popup

Click the toolbar icon on any page and hit Scan. Quick and Deep mirror the same two modes on the web app's scan form: Quick runs the fast header/TLS/content family checks, Deep also crawls linked pages on the same host. Which check families run is controlled from the extension's own Settings, independent of your web app defaults, so you can keep the popup fast day-to-day and still reach every check when you want it.

A scan started from the popup counts against the same daily limit as any other scan on your account and shows up in your regular scan history: there's no separate extension-only history to lose track of.

Reading a result in the popup

The popup is not a summary you then go elsewhere to act on. Three things happen without leaving it:

  • Expand any finding for where it was found, why it matters, the numbered fix steps, and the reference links. The API already returned all of that; the popup used to render only the title and description and drop the rest.
  • Export a saved scan as PDF, SARIF, Markdown, or JSON. It calls the same GET /api/v3/history/[id]/report the dashboard and CI use, so the file is byte-identical rather than a second formatter that can drift.
  • Per-site history, where each row shows the delta against the next-older scan of the same URL (fewer or more findings) and carries an inline rescan button.

The on-page card

When enabled, a small card can appear in the corner of a page reporting one of two things: this host has been scanned before (and what its last result was), or it hasn't and you can scan it now. Site alerts are entirely separate from auto-scan below: this is about surfacing information, not triggering a scan on its own.

  • Doesn't repeat itself: once you've seen the card for a host, it stays quiet on later page loads and tab switches to that same host for 24 hours, unless a new scan actually changes the result.
  • Not this site: mutes the card for that host (or a URL pattern) permanently, from the card itself or Settings > Site Alerts.
  • Snooze 24h: a temporary, self-expiring version of the same mute for one host.
  • The known-result and scan-prompt halves of the card are each controlled by their own toggle in Settings, so you can keep one and turn off the other.
The on-page card in the corner of a checkout page, showing a danger score of 3, one high and one medium finding, when it was last scanned, and Snooze 24h, Not this site and Turn off controls
The card on a host scanned three days ago: the danger score, the severity counts, and the three ways to make it stop.

Auto-scan modes

Off by default. Three modes decide when a scan fires without you clicking anything:

On page load
Scans a host the first time you land on it, throttled per host so repeat navigation on the same site doesn't re-trigger.
On tab focus
Scans when you switch to a tab showing a host that hasn't been checked recently, not on every load.
On URL change
Follows SPA/client-side navigation within a site, not just full page loads.

A whitelist/blacklist and a global pause are available in Settings for hosts or stretches of time you never want auto-scanned, regardless of mode.

The extension settings page on Site Alerts, showing the two on-page card toggles above an auto-scan trigger picker with Off, On tab focus, On page load and On URL change
Settings > Site Alerts. The card toggles and the auto-scan trigger are separate controls: the card can stay on with auto-scan off.

Desktop notifications

Any scan that wasn't started from the popup finishes with a desktop notification carrying the finding count and the worst issue, because a background scan you never look at is not worth running. Clicking it opens the full report. The severity threshold that decides whether a result is worth notifying you about is in Settings, so a site with three info-level findings can stay quiet.

A desktop notification reading: 3 findings on shop.example.com, Missing Content-Security-Policy header +2 more
A finished background scan. The notification names the host and the worst finding, and opens the report when clicked.

Pages you're signed into

If the extension detects a session cookie on the current page (a heuristic against well-known cookie names, not a guarantee), the reputation card says so and links to authenticated scanning on the web app instead of prompting you to scan it directly. A regular scan always checks a site the way a logged-out visitor sees it (headers, TLS, cookies, DNS), which is a fundamentally different (and mostly incompatible) job from reading your actual authenticated DOM, so the extension doesn't try to capture and scan the signed-in page itself. For that, use the "Sign in first" option on the web app's scan form, which drives a real login for the scan.

Settings

Open the full settings page from the popup's gear icon. Everything above (check families, the port and service sweep, auto-scan mode and throttle, notification threshold, card position, mute lists, and theme) lives there and is stored locally in the browser, not on your account, so it's per-install rather than per-user. The sweep is one toggle now, not a per-service list: the old per-service panel serialised a probes array the API had already stopped reading, so configuring it did nothing.

Permissions and privacy

  • Runs on every page (content script) so the on-page card and auto-scan work without a click first. It reads the page's URL and a same-origin session cookie name check; it does not read page content into anything that leaves your browser.
  • Your API key is stored in the browser's local extension storage, sent only to VulnRadar's API, and never to any other origin.
  • Nothing is scanned without either you clicking Scan or an auto-scan mode you turned on yourself.
  • Fully open source, same license as the rest of VulnRadar. Read the extension's source in the GitHub repository.
Questions about a specific permission prompt? Ask in the contact form.
Browser Extension for Chrome and Firefox | VulnRadar