Repository navigation
Releases
Update Scanner Reborn 5.3.0 build 119
5.3.0 build 119 - 2026-11-11
This is the release that follows 5.1.6. It contains everything from 5.1.7 to 5.3.0.119, which were test builds and were not published on their own. The first part below is what is new since the first 5.3.0 notes (further down); everything in those notes still applies.
Newer scanning: Automatic now steps up through Standard, Dynamic and Popup
- Automatic works in three stages and stops as soon as one succeeds. Standard first (up to 10 seconds, retried while the page is still loading or recovering), then Dynamic (its own time limit, see the next item), then Popup as the last fallback. A scan that passes at any stage is saved and the later stages are never started.
- Automatic remembers what a page needed. A link that needed Popup goes straight to Popup next time, a link that needed Dynamic starts at Dynamic (and moves on to Popup only if that fails). Choosing Automatic again, or changing the address, starts the check over.
- A mode you pick by hand is still the mode that is used, and Lock still means no other kind of scan is ever tried. Only Automatic moves between modes.
- Standard no longer accepts a page that is only its outline. A Standard capture is checked for pictures that are only placeholders, lists a script fills in afterwards, and tall areas that hold nothing, and Automatic moves on to Dynamic when the page is not really there. A hand-picked Standard never changes mode; it keeps trying inside its 10 seconds and, if the page is still empty, keeps your last good copy instead of saving an empty page.
- A site that is down cannot push a page to a slower mode.
New: Maximum Dynamic Scan Time (Global Settings)
- A new slider in Global Settings, Dynamic scanning, sets the longest a Dynamic scan may take: 20 to 45 seconds in 1-second steps, 25 by default. It applies to Dynamic scans and to the Dynamic stage of Automatic.
- A Dynamic scan finishes early as soon as the page is complete; the limit is only how long it may keep trying. In Automatic, a Dynamic scan that runs out of time moves on to Popup.
Improved: Dynamic scanning catches far more lazy-loaded pages
- The page is scrolled by the extension itself, row by row at a human pace, instead of by the page's own timers, which Firefox slows down in a background tab. Galleries that load pictures as you scroll now load completely in the background tab far more often, so Popup is needed much less often. (Popup is still the faster scan on some sites.)
- Pictures that are in the page but have not been decoded or painted yet are noticed and waited for, and a scan that reached the bottom is no longer failed only because the clock ran out during the final clean-up.
- Zoom is set to 100% for the scan and your own zoom is put back afterwards.
New: Include, Ignore and Exclude: "Select parts"
The Ignore parts... button is now Select parts..., and its window has three choices for what Add & Next does:
| Choice | What it does |
|---|---|
| Include | Only the selected regions count as changes. The rest of the page stays visible but is not compared. Use it when one area (a list of releases, a price) is the only thing you care about. |
| Ignore (default) | The selected part is left out of the comparison but stays visible. |
| Exclude | Ignored too, and removed from the saved view; can be copied as a uBlock Origin filter. |
- Exclude wins over Ignore, and Ignore wins over Include when parts overlap. With no Include rule the whole page counts as before.
- With an Include rule, changes to the page's head (title, meta tags) do not count.
- Page Settings, Selected parts (it was "Ignored details") lists every Include, Ignore and Exclude rule; each has Remove and a button that moves it to the next kind (an Include rule offers Make ignore, an Ignore rule Make exclude, an Exclude rule Make include). Add a rule by hand has an Include choice too.
- Include, Ignore and Exclude rules are carried by Backup Pages and Restore.
- The tuning bar's All changed preview leaves ignored, excluded and not-included parts alone, so it shows only the blocks that really count.
Improved: the picker on ad-heavy sites
- While you select, empty click-catching layers are looked through, so the card or picture under an advert's see-through overlay can be selected. Nothing on the saved page is changed.
- Position-only rules are refused ("Unstable element"). Ad networks rebuild those elements on every load, and such a rule could later hide an unrelated button. Use a stable part, the automatic click-layer protection, or block the script host.
- Clicking inside a saved part no longer removes the rule around it, and the same rule cannot be saved twice.
- The live-page picker works on any page of a monitored site, not only the exact address you scan. If the site is not monitored the window says to add a page of it first.
- If an advert opens a top-layer pop-up over the picker, the picker puts itself back on top. A narrow frame no longer clips the picker window.
- Select a visible ad / iframe: the real, visible advert can be selected without running its click.
- Block pop-ups opened by scripts and Make hidden click layers click-through are stricter: only empty, viewport-covering layers are touched, never visible dialogs, video players or the site's own controls.
New: block a script host or third-party frames on a site (optional)
- In the live picker, a third-party script host (an exact host name, for example
ads.example.net, no wildcards and never the site itself) can be blocked on that site, and third-party frames can be blocked on a site as a whole (off by default, because some sites need embeds for their videos). The window warns about this. - They work only for sites you monitor, only when Hide excluded parts on sites is on, and only for the hosts you chose. A request the rule does not match is always allowed. Reload the website after saving. Remove a rule under Saved parts to unblock it.
- This needs the webRequest and webRequestBlocking permissions (new in this release). Nothing is read, recorded or sent anywhere.
New: websites that are down, and the three-hour recheck
- A failed scan no longer replaces your last good copy. The saved page and its changes are kept.
- The first failure shows grey with the reason (the sidebar hover says when the next try is). One confirmation scan runs three hours later, not on the normal interval. If the site is still unavailable, the link turns red, you get a "website still down" notification once, and the toolbar badge shows the number of down sites with a red !.
- Cloudflare error pages are recognised (error 521 and the like, even with a blank title or extra markup). An ordinary article that only talks about Cloudflare or HTTP errors is not mistaken for one.
- A good scan clears the status and a down site never counts toward promoting a link to a slower scan mode.
- A capture that was rejected as incomplete is not counted as a finished scan: the previous copy stays, and a folder scan no longer reports a misleading "0 changes".
New: Delete all pages (Folder and Global Settings)
- Folder Settings has Delete all pages in this folder... and Global Settings has Delete all pages... (a request from users). It deletes the pages in one go and can include the folders inside.
- Two guards: read what will be deleted and tick I understand that this cannot be undone, then type DELETE (DELETE ALL for the whole list) before the Delete button works. Enter never deletes by itself, progress is shown page by page, and Cancel or Escape leaves everything as it was.
- Make a Backup Pages file first; a deleted page cannot be brought back except from a backup.
Changed: scan order and counts
- Right-click Scan goes next, for a page and for a whole folder, ahead of the work already waiting. If that page is already being scanned, the running scan finishes and a fresh scan of it follows, using the latest settings.
- Numbers like
100% (34)or21.3K (1,200)inside one highlighted element are treated as numbers with "Ignore changes to numbers". - The toolbar popup lists sites that are down as changed.
For testers: diagnostic tools (hidden)
Click the Update Scanner arrow in the toolbar popup header seven times (each within 2 seconds) to show Test: capture each Dynamic pass and View captured passes. They are off by default, hiding them switches capturing off, and the captures stay inside the browser (a viewer page; nothing is uploaded). A capture adds about 15 seconds to the Dynamic limit.
5.3.0 (first notes; everything here still applies)
This release also contains everything from 5.1.7 to 5.2.0, which were test builds and were not published on their own.
New: count changed blocks (posts, items) instead of characters
Listing pages (forums, blogs, shops, "latest uploads" pages) show their content as repeated blocks: a post with a title, a date, a category, a picture and a few lines of text. Counted by characters, ONE new post can add up to hundreds of changed characters, so a page with two new posts looks like a huge change. Counted by blocks, it is two changes.
- New setting "Count each changed block (post, item) as one change" in Page Settings, Folder Settings and Global Settings (off by default, so every existing link behaves exactly as before). A post, list item or table row with anything changed in it counts as ONE change, however much of it changed.
- What is a block: the outermost repeated element of the page (every post of a listing, every item of a list, every row of a table) that is still post-sized (about 1500 characters or less). Text with no repeated element around it counts as one block ...
v5.1.6
5.1.6 - 2026-11-03
New: jump from change to change
- Next / Previous buttons in the Changes view. On a long page you no longer have to scroll to find what changed: press Next or Previous (under the view dropdown) and the page scrolls to the next highlighted change and flashes it with an orange outline. The counter shows where you are ("Change 3 of 12").
- One change, one stop. Highlights that sit close together, such as the words of one edited sentence, count as a single change.
- Sticky parts of a page are not stops. Parts that stay on the screen while you scroll (a sticky sidebar such as "On this page", a fixed bar) are still highlighted, but Next and Previous skip them, so the count does not change as you move.
- Works from where you are. If you scrolled by hand, Next goes to the first change below what you are reading and Previous to the one above. At the last change Next goes back to the first, and Previous from the first goes to the last.
- Keyboard: press N for the next change and P for the previous one.
- The buttons only appear in the Changes view when something is highlighted.
v5.1.5
5.1.5 - 2026-11-03
This release also contains everything from 5.1.2 and 5.1.4, which were not published on their own (see below).
Changes
- Help opens the Update Scanner Reborn issues page (github.com/molitar/update-scanner-reborn/issues) instead of the website of the original Update Scanner.
- The toolbar popup now waits for a new tab to be created before it closes, so Global Settings and New always open their tab beside the current one.
v5.1.4
5.1.4 - 2026-10-14
Faster scans
Scans are quicker without capturing less. Typical times:
| Scan mode | Typical time |
|---|---|
| Standard (first check) | about 10 seconds, then a plain fetch of about 2 seconds on every later scan |
| Dynamic | about 15 seconds when two sweeps are enough, longer (up to about 25 to 30) when a third and fourth sweep are needed |
| Popup | about 30 seconds |
- Dynamic now typically finishes in about 15 seconds, after two passes, and only takes longer when a site needs a third or fourth pass. Overall scan times are noticeably shorter. Dynamic and Popup scan in up to four passes: The page is swept from top to bottom a little slower each time (about 4, 5 and 6 seconds). After the second sweep the scanner checks whether every image has loaded; if so, the third sweep is skipped. If images are still missing after the third sweep, one slower fourth sweep (8 seconds) follows, and only then is the page judged to need a deeper mode.
- Short waits. The page starts to be scrolled 1 second after it is ready, so lazy-loaded images start loading sooner. The check that every image has loaded no longer waits on timers that a background tab slows down. When everything is loaded the scan goes back to the top, settles for 2 seconds and closes, without a long final wait.
- A slow ad can no longer hold a scan back for 20 or more seconds.
- The Standard check is a quick look, never more than 10 seconds. The check that decides whether the fast Standard scan is enough opens the page, makes one quick pass over it and reads it, without building a saved copy.
New: Automatic scan mode and a Lock switch
- Scan mode now starts with Automatic (recommended), followed by Standard, Dynamic and Popup, in Page Settings, Folder Settings and Global Settings.
- Automatic works out the best scan mode once. On the first scan it starts with Standard, and if the page is not fully captured it moves to Dynamic, then Popup. After that it keeps that mode, and the dropdown shows what it found, so a site that is down for maintenance can never push a page up to a deeper mode later.
- Automatic settles on sites that block a plain fetch. A site that answers the Standard scan with "Forbidden" or a Cloudflare check goes to Dynamic straight away and is checked against Popup.
- A mode you pick by hand is the mode that is used. Dynamic and Popup are never checked or changed. A hand-picked Standard gets the same 10-second quick look once and only writes what it finds in the log; it never changes the mode.
- Lock scan mode is your own choice, picking a mode does not turn it on. A locked mode never changes for any reason, is never checked, and no other kind of scan is tried even when scans keep failing (you get a normal error instead).
- Reset to Automatic. For one link, choose Automatic or turn the lock off; it starts over and is checked once, on its next scan. For a folder or Global Settings, choose Automatic and tick the box under Scan mode to reset all existing links.
- In a folder or in Global Settings, the scan mode is also the default for new links.
Better settings
- Simpler Folder Settings and Global Settings. Each "Apply to existing links" box sits directly under the setting it covers (Autoscan, Change Threshold and numbers, Scan mode and lock, Notifications) with a label that says exactly what it changes. They all start unchecked, so nothing on your existing links is changed unless you tick it. The separate block of boxes at the bottom is gone.
- Sort your folders and links. Global Settings and Folder Settings can sort everything alphabetically (A to Z, numbers in order). List folders before links is no longer ticked by default, and becomes selectable once you tick the sort option.
- Settings open in their own tab, to the right of the tab you are on. They no longer replace the page you were looking at, the tab closes when you are done, and you land back on the tab you came from instead of one that is busy scanning.
- Pages open where you expect. From the sidebar or the toolbar popup, a page you click opens in the same tab when you are already looking at an Update Scanner page, and in a new tab right next to the current one when you are on any other site, so the site you were reading is never replaced.
More accurate captures
- The page header says how it was scanned, for example "This page was last scanned today using dynamic at 21:36." It shows the mode that really did the scan.
- Standard must show a whole page. The Standard check looks at the saved HTML the way the viewer shows it (without scripts). If most of its text stays invisible until the site's scripts reveal it, if it shows less than one page, or if far fewer words show than on the live page, Automatic moves the link to a deeper mode.
- Pictures in saved pages keep their original size and detail. When a site refuses direct downloads, the scanner fetches the picture's own file the way the page does, instead of redrawing it, so covers are saved whole.
- A "verify you are human" check no longer gets the Popup scan window closed on you. If a site (for example one behind Cloudflare) shows its checkbox in the Popup scan, the window is brought to the front, a notification tells you, and the scan waits up to 90 seconds for you to tick it.
- Dynamic and Popup scans no longer jump to a deeper mode because slow ads did not load.
- Loading placeholders alone no longer force Dynamic. Grey shimmering bars (like the commit details on a GitHub page) are only noted in the log; the page's real text decides.
v5.1.2
5.1.2 - 2026-10-13
New: sort your folders and links
- Global Settings has a new Sorting section: tick Sort all folders and links alphabetically (A to Z) and press OK to sort the whole list in one go.
- Folder Settings has the same option for a single folder: Sort this folder and the folders inside it alphabetically (A to Z).
- Sorting ignores upper/lower case and puts numbers in order (Video 2 comes before Video 10). List folders before links is on by default; untick it to mix folders and links together.
- It only reorders; nothing is renamed or removed, and folders that are already in order are left untouched.
v5.1.1
5.1.1 - 2026-10-13
A small fix on top of 5.1.0. Everything in the 5.1.0 notes still applies.
Fixed
- Folder Settings now has the Autoscan interval, like Global Settings. Links added to the folder start with that interval. Tick "Apply to all existing links" and "Apply Autoscan interval to existing links" to update the links already inside the folder.
v5.1.0
5.1.0 - 2026-10-13
Everything added since 5.0.0.
New: backups that survive a reinstall
- Your pages are copied automatically into your bookmarks, in an
Update Scanner Reborn folder under Other Bookmarks. Your folders and pages
appear as ordinary bookmarks, and a "Backup data" folder holds a full backup
(every page setting included) for each of the last 14 days. The browser's own
daily bookmark backups protect that folder as well. - A copy of your pages is saved before every restore, because restoring replaces
all pages. - If the extension's storage is ever empty (for example after the browser gives
the add-on a new internal ID), it restores itself from the newest bookmark
backup the next time it starts. - The Restore page lists the automatic backups, so you can go back to an earlier
day as well as restore from a file. - If the page list suddenly shrinks to less than half, the earlier backup is kept
untouched and the smaller one is saved next to it.
New: only what you can see counts
- Rendered scans (Dynamic and Popup) mark the parts of a page that are not
showing when it is captured: other carousel slides, closed menus and sections,
off-screen text and screen-reader-only text. Those parts are ignored when
comparing and in the Changes view. The page title, other head metadata,
<noscript>and<template>content are ignored too. Rescan a page once so
its snapshot carries the marks. - Script and style code (analytics scripts, inlined CSS) is left out of the diff,
so it never shows up as a change. The page still displays as captured. - Rendered scans are taken at the same zoom every time, and each snapshot records
the zoom and window width it was captured at. If the old and new pages were
captured differently, the Changes view says so.
New: change size and thresholds
- Finer change-threshold steps: every 10 characters up to 200, every 20 up to
300, every 25 up to 500, then every 100 up to 1500 (the highest used to be 500).
Existing page thresholds are unchanged. - A change is measured in letters and digits of the visible text that changed.
Spaces, punctuation and decoration such as[ENG]or|no longer count. - With "ignore numbers" on, a highlight that is only a number, or a number with a
%orK/M/Bsuffix (ratings and view counts such as72%or8.1K), is
not counted as a change. - A highlighted date or time ("October 12, 2026", "7 hours ago", "10:30 PM")
counts as a single change however long it is written, instead of one change
per character. - Folder-wide change threshold and number handling, applied to every link in the
folder (including subfolders) when you choose. - Page settings show how big the changes on recent scans were, with a one-click
suggested threshold that would have ignored them all. - "View Changes" shows how much changed and how far that is over or under the
threshold (for example "56 changed, 16 over the 40 threshold"). Hover it for
how to read the numbers, what is normal for the page, and the text that was
highlighted. Hover a link in the sidebar to see its last change and what is
typical.
Improved: the Changes view
- Unchanged words are no longer highlighted: random IDs that pages regenerate on
every load are ignored, a highlight on text that appears equally often in the
old and new page is dropped, and when no text changed the page is shown without
highlights. - Inlined page CSS is no longer counted as page text when comparing.
Fixed: Folder Settings
- Folder Settings now has the Autoscan interval, like Global Settings. Links added to the folder start with it, and "Apply Autoscan interval to existing links" (with "Apply to all existing links") updates the links already inside.
Upgrade notes
- New permission:
bookmarks, used only for the automatic backup above.
Firefox asks you to approve it when you update. - Install the new version over the existing one. Do not remove the add-on first:
removing it deletes its stored pages. - Rescan each page once so its snapshot carries the hidden-part marks.
v5.0.0
Update Scanner Reborn v5.0.0
Initial public release of Update Scanner Reborn, a modernized continuation of the original Firefox Update Scanner extension.
Features
- Intelligent scanning with Standard, Dynamic, and Popup rendering modes.
- Automatic detection and remembered scanning preferences.
- Improved snapshots for JavaScript-heavy websites and images.
- Global, folder-level, and per-website settings.
- Improved visual change highlighting.
- Manual scan priority with automatic queue continuation.
- Background scanning without stealing browser focus.
- Live sidebar scanning indicators.
- Configurable desktop notifications and audio alerts.
- Firefox 140+ support.
Installation
Recommended: Install through Mozilla Firefox Add-ons once approved.
https://addons.mozilla.org/en-US/firefox/addon/update-scanner-reborn-v5-0-0/
The attached XPI is the packaged build and may require Mozilla signing before installation in standard Firefox.
Downloads
- Update_Scanner_Reborn_v5.0.0.xpi — Firefox extension package.
- Update_Scanner_Reborn_v5.0.0_Source.zip — Packaged extension source.
The source is also available directly in this GitHub repository.
Credits and licensing
Based on the original Update Scanner by sneakypete81 and contributors.
Licensed under GNU GPL v3. See LICENSE, NOTICE.md, and THIRD_PARTY_LICENSES.