Powerful permissions, clearly explained.
WardenOne protects across active browser surfaces: pages, redirects, downloads, trackers, cookies, sessions, and installed extensions. This page shows why each permission exists and where its use stops.
What each browser permission does.
This ledger groups permissions by the job they perform — naming the permission, why it exists, and where its reach stops. Eighteen are requested at install; browser-history access is optional and requested only if you turn on history clearing in Forget Me.
<all_urls>RequiredGives WardenOne coverage on the sites you visit so phishing, tracker, download, redirect, and session protections do not miss pages. This one scope also covers the outside services listed below and any filter-list address you choose; the manifest does not present a partial list of those hosts as extra permissions.
declarativeNetRequestCoreLets Chrome enforce fast block rules for ads, trackers, phishing, malware, scam, and IP-grabber domains.
activeTabCurrent tabAllows a popup, shortcut or right-click action to receive temporary access to the current page.
scriptingPage shieldApplies page-level protection such as overlay cleanup, form trap detection, token checks, and trusted-site adjustments.
webRequestDetectionDetects risky redirects, suspicious beacons, token exfiltration attempts, and download chains that static lists cannot fully explain.
webNavigationNavigationWatches page movement so WardenOne can catch redirect chains, fake login hops, popups, and risky landings.
storageLocalSaves settings, list metadata, activity, allowlists, trusted sites, API keys, local learning data, and installed-extension change/review snapshots.
alarmsLocalSchedules list refreshes, startup checks, extension-inventory reconciliation, retry checks, provider cache cleanup, and maintenance tasks.
notificationsAlertsShows local warnings when a threat is blocked, a risky extension changes, or a security check finishes.
historyOptionalClears a site's entries from browser history when you explicitly enable that part of Forget Me. WardenOne asks Chrome for this access when you turn the switch on.
browsingDataCleanupPowers privacy cleaning, emergency sign-out, site data clearing, cache cleanup, and session reset actions.
cookiesCleanupSupports strict cookie controls, session safety checks, Forget Me, and emergency logout cleanup.
managementReviewBuilds the local Extension Security Centre from exact IDs, versions, enabled state, install type, Chrome permission warnings and access changes. The centre can disable an extension or ask Chrome to confirm removal only after you press that extension's button.
contentSettingsSite controlReads or adjusts camera, microphone, notifications, cookies, JavaScript, images, and other site permissions.
downloadsGuardChecks risky downloads, pauses suspicious files for review, and lets you cancel or continue with context.
tabsTabsReads tab URLs and titles for duplicate-tab scanning and Memory Shield, including when Chrome restricts WardenOne’s site access.
declarativeNetRequestFeedbackFilteringLets the Network Logger say which filter rule blocked a request, not just that something did. Chrome only joins a rule to an individual request for an extension loaded unpacked; a Store install gets the weaker form, which reports rules by tab and time, and the Logger says so on the page rather than guessing an attribution.
contextMenusInterfaceAdds WardenOne’s own entries to the right-click menu — check this link, scan this image, block this site, sleep this tab.
offscreenInterfaceOpens a hidden extension page for enabled notification sounds and link copying requested from WardenOne controls. It closes after a short idle period.
clipboardWriteCopy controlsCopies the clean link you requested from the extension's own document, so the website cannot replace it with another value.
What can leave the browser.
Every address WardenOne contacts outside your browser, split by whether it happens on its own or only after you switch something on. Nothing is ever sent to WardenOne’s developer — there is no WardenOne server to send it to.
Downloads the public ad, tracker, malware and phishing lists from their maintainers once a day, with a few retries if a download fails. Each list is fetched only while the protection that uses it is on. These are plain file downloads without cookies: nothing about the pages you visit is sent, and the list host sees an ordinary request from your IP address. Download Guard's malware address and file-hash lists arrive the same way and are matched on this device. Every list and its exact address is published in the source inventory.
Community-maintained lists help identify advertisements, trackers, phishing sites, scams and malicious domains. Their maintainers retain ownership and credit. WardenOne records published licenses and separate written permissions and supplies attribution and license notices in the extension. Specific scope exceptions for aggregate feeds containing OpenPhish-derived data are explained in the source review.
Update Guardian asks your browser maker for its latest public Stable release when you open the popup or press Check again. Only the source for your browser is contacted. No browsing history, page address, settings or cookies are sent; the maker sees an ordinary request from your IP address. The result is reused for up to six hours while the browser stays open; a restart asks again.
For an official GitHub ZIP, WardenOne asks GitHub for the latest complete release. If its build differs, it asks whether your public build commit is an ancestor. GitHub sees your IP address and, for that comparison, the two public commit IDs. No browsing history, page address, settings or cookies are sent. Results are reused for up to six hours; failed checks retry after ten minutes. Settings → About WardenOne has Check again. Source and Store builds are not checked.
To notice when a script a site relies on has been quietly swapped, WardenOne re-requests up to 14 third-party scripts a page has already loaded, hashes them on your device and compares the hash with the last copy it saw. Each request goes to a host that page just used, without cookies, so no one learns anything new. Scripts from the site itself, and allowlisted sites, are skipped. On by default; switch it off with Script drift guard.
Only while a Twitch page is open. Twitch ad blocking asks Twitch for the stream’s playback token the way the player does, carrying the same Authorization header the Twitch page is already using — sent back to Twitch only, never stored, and without cookies. Twitch rewind asks whether the channel has an in-progress recording of the broadcast, and sends the channel name with no credentials at all.
Used only when you click Check breach history. Sends the site's registrable domain to look up that site's public breach record — not your account, and not any password.
Used only when you type a password into Check password exposure and press the button. The password is hashed on your device and only the first five characters of the hash are sent; the service replies with every suffix in that bucket and the match is made locally. The password and the full hash never leave your machine, the request carries no cookies and no referrer, and the answer is written nowhere. WardenOne does not read what you type into web pages.
Used for registration age and reputation clues when a new or suspicious domain asks for passwords or serves risky downloads.
Optional URL reputation checks when you provide and enable your own API key. Once enabled it is asked on its own about every page you open, risky link, sensitive form destination and download, and receives the address as scheme, host and path — never the query string.
Optional URL and file-hash reputation. WardenOne can check a hash without uploading the downloaded file.
OpenPhish is on by default: its public feed is downloaded and matched locally without submitting your browsing addresses. PhishTank URL lookups remain opt-in and require your own key.
Optional malware URL and abusive IP checks for suspicious servers, raw IP addresses, and risky download sources.
Used only when you paste an extension’s ID or store link into the extension checker. Asks for that listing’s public page, without cookies, to learn whether it still exists and what it is called. Sends the extension’s ID and nothing about you.
Used only when you subscribe to a filter list in My Rules and when you press its update button — never on its own. The list is fetched from the exact https:// address you typed, without cookies, so that host sees an ordinary request from your IP address and nothing else.
Used only when you open the network test page and start it. It tries to load a favicon from two adult sites, one malware test host and two control sites, and sees which are blocked. Nothing is shown, but these are real requests: they can appear in DNS logs, in router, school or workplace monitoring, and in family-filter reports. On a network where that matters, don’t run it.
Where you control the access.
The popup keeps the operational controls visible, so broad browser protection can still be softened per site or strengthened per feature.
Protective by default, adjustable when needed.
Use the popup to tune protections, set provider keys, trust sites, allow download sources, review extensions, or clean browser data.
Per-site adjustments
Allowlist a site, trust a download source, or adjust camera, microphone, notification, cookie, and JavaScript settings.
Visible review tools
Check domain age, breach history, exposed tokens, startup risk, installed extensions, and site permission settings.
Cleanup when you ask
Forget Me, emergency logout, cookie cleanup, and browser-data cleaning are user-started actions.