What’s new
WebTrustScore website & app changelog
Updates to the website, the app, and the infrastructure behind them — new features, design changes, and fixes. This is product history; changes to how the trust score itself is computed are tracked separately on the Model history page.
v1.15.1 — 2026-08-20
The bulk list holds its shape while a run is going. The word "queued" used to appear inside a row, between the status chips and the numbers — and because every row shares its column widths with the rest of the list, one waiting row pushed every row’s columns apart until the run finished. A row that is waiting or being scored now shows a small bold badge floating over it instead, with the row softly washed out behind it, so scores, chips, and buttons stay exactly where they are while the work happens. Removing or reordering rows during a run works as before; nothing about scoring changed.
v1.15.0 — 2026-08-15
Your bulk list no longer travels in the web address. It used to be written into the address bar every time you sorted, reordered or ran it, which meant your working list was visible to anything that reads a URL without you choosing it — browser history, page analytics, server logs. It is now kept with your account instead, and the address bar stays clean. Sharing a shortlist still works, and works better: "Copy share link" creates a link that contains nothing but a random code, so the names live with your account rather than in the link you send. A shared link carries the names only — never your scores — expires after 90 days, and creating a new one replaces the old one, which is how you cancel a link you have already sent. Following someone else’s link fills the box with their names and scores nothing, so it never uses up your own allowance. Your list is also still kept on this device as a fallback. Three supporting fixes came with it: if saving to your account fails you are now told, instead of finding out later on another device; closing the tab no longer loses the last few seconds of changes; and a single failed connection when the page opens no longer quietly switches off saving for the rest of your visit.
v1.14.2 — 2026-08-15
The bulk list no longer slides sideways. Long website names used to push the row wider than the space it had, so reading one meant scrolling the list across — and the space was tighter than it looked: after the side menu and the padding around it, only about fifteen characters of a name fitted before that started, which is shorter than a great many real names. The side menu is narrower now and the spacing around the list has been tightened, which roughly doubles what fits. Beyond that the list watches its own width: when a name is too long for everything to sit on one line, that row splits onto two lines and the name keeps a full line to itself, rather than the list scrolling. Rows go back to sharing aligned columns as soon as they fit again, so a list of ordinary names looks exactly as it did. Names are still never shortened and never broken across lines. On a phone a name wider than the screen still scrolls within the list itself — at that size there is no arrangement that fits it.
v1.14.1 — 2026-08-15
Two changes to the bulk list. You can now add more websites while a run is still going, instead of waiting for it to finish — the ones you add join the run already in progress and are checked alongside it, and everything already on screen is left exactly as it is. Re-checking is still held back until the current run finishes, since that replaces scores that are being worked on. Separately, removing a row now asks first, and so does Start over — which says how many websites it is about to clear, and clears the saved copy of your list along with them.
v1.14.0 — 2026-08-15
Bulk checking now has everything the DomainValueScore version has. Your list is remembered against your account, so it is still there when you sign in on another computer — the same scores you left, with their dates, and re-checking stays something you ask for rather than something that happens on its own. Website names are now links that open the site in a new tab, opening in a new tab so your list stays where it is; a name always goes to exactly the site it displays, and if it cannot, it is shown as plain text instead. Please treat that as a convenience, not an endorsement: a name in this list may be one this tool has just flagged, and the score beside it is the thing to read first. Alongside those, three shared improvements: names stay on one line and are never shortened, the star is gone and the ± column it controlled now sets what you are comparing against, and the row itself opens its detail.
v1.13.2 — 2026-08-15
A small change to how a bulk row opens. The arrow beside each website name is now the control that opens its detail, which is what makes the row reachable from the keyboard; clicking anywhere else on the row still opens it too, exactly as before. Website names here stay plain text and are not links: this list routinely holds sites that have just been flagged for malware or phishing, and a one-click route to them from the tool that flagged them is not a default worth having.
v1.13.1 — 2026-08-15
In bulk scoring, a result now opens by clicking the row itself. Anywhere on a row that is not a button, a link or a control will open it — the score, the band, the empty space between them — and clicking again closes it. The separate Details control on every row is gone, since the row it sat on now does the same job with more of it to aim at; the website name is the control that carries it, so this still works from the keyboard exactly as before. Clicking inside a result you have opened does not close it, selecting a name to copy it does not close it, and the Report link, the comparison tickbox, the baseline star and the reordering buttons all still do only their own job. Rows that have nothing to show yet — still scanning, or a name that could not be scored — are not clickable and do not pretend to be. Also fixed: a row still being scanned had its buttons sitting one column to the left of every other row.
v1.13.0 — 2026-08-13
Bulk scoring gains full reports in place, and names you can add as you go. Every scored row now has a Details control that opens the whole report where it sits — each category with its score, its weight, how much of its evidence could be collected, the evidence itself, any notes, and the reason a score was capped. Several can be open at once, which is the point: two full reports side by side is the one thing a link to a separate page cannot give you. Nothing is fetched to open them, so they appear instantly. You can also add more websites to a list that already has results without starting again — only the new ones are scored, and your existing scores, your order, your chosen baseline and any open reports all stay exactly as they were. Removing a row was already there. The list still holds twenty-five.
v1.12.1 — 2026-08-13
Bulk scoring moved to where your other account tools live. It now sits in your account, directly above Scan history, instead of in the Tools menu at the top of the site. Every signed-in member can see the section — it is no longer hidden from anyone with an account — and opening it either shows the tool, if your plan includes it, or explains what it does and what it costs if it does not. The page it used to live on still works if you have it bookmarked.
v1.12.0 — 2026-08-12
Bulk scoring, for members. Paste up to twenty-five websites — one per line, or straight out of a spreadsheet column — and get every score in one run, on one screen. Names that have been scored recently come back immediately; the rest fill in as they finish, so the list is useful before it is complete. Sort by score, by name, by any single one of the seven categories, or by what needs a look first; pin any row as the baseline and every other row shows how far above or below it sits; check two to four rows to see their categories side by side with the best value in each row marked. Drag the list into whatever order you want and it stays that way, arriving results appended rather than shuffled in — and the same reordering works from the keyboard, or from the up and down buttons on each row. The whole run exports to CSV with every category, the model version, and a link to each report, or copies as a table to paste elsewhere. The list lives in the page address, so a shortlist can be shared as a link. Scoring one website stays free, as it always has been; scoring a list at once is part of the paid plan.
v1.11.0 — 2026-08-06
A Methodology page, and a simpler top menu. The full method behind the score now lives on its own page, in all eight languages: what the seven weighted categories measure and how much each one counts, which findings cap a score outright and at what number, why a domain with no working site gets no score at all, how confidence is reported separately and can only lower the ceiling, what happens when a source cannot be reached, and what the version stamped on every report means. The top menu drops its “Model” link — what it pointed to is covered in full on the new page, which sits in its place — and the section itself has not gone anywhere: the “How it works” button on the home page still opens it. The model history now also shows the date each version went live, so a score can be read against when its model was in force.
v1.10.2 — 2026-08-05
The live preview on the widget page now grows to fit what it is showing. It used to be a fixed height, so as soon as a score appeared the result sat below a scrollbar — the one thing the preview exists to demonstrate was the part you had to scroll to find. It now expands to the full height of the result, and the empty space between the introduction and the Options panel has been reduced so the configurator and the preview are both in view when the page opens. The embed code itself is unchanged.
v1.10.1 — 2026-08-05
Clicking “Get score” more than once no longer starts more than one check. The button and the address box now switch off the instant you click, instead of staying live while the request is on its way — so an impatient second or third click can no longer produce several results for the same site. The server now also settles, in one atomic step, which request starts the check when several arrive together, and points the rest at the check already running. Separately, Notable Websites has moved out of the top menu: it is still in the footer, and there is now a link to it on the Reading your report page, right under the score bands it illustrates.
v1.10.0 — 2026-07-31
The Chrome extension is out. WebTrustScore is now on the Chrome Web Store: open the side panel, score the tab you are on or any domain you type, and read the same 0–1000 trust score this site gives — with a local, never-transmitted read of what your own browser sees on the page in front of you. A new Extension page explains what it does, what it sends and what it never sends, and links straight to the listing, in all eight languages.
v1.9.2 — 2026-07-13
Scan hardening. The HTML link and text parsers used on every scanned page were rewritten to run in linear time, removing a denial-of-service risk where a maliciously crafted page could make a scan hang. Behavior is unchanged for normal pages; scores are unaffected.
v1.9.1 — 2026-07-11
Scan results now expire with the model. A cached scan is reused only if it was produced by the current scoring-model and evidence-collection versions, so a scoring or detection fix takes effect immediately after release instead of being masked by up to a week of cached results.
v1.9.0 — 2026-07-11
Reading your report. A new report guide defines every line item on a report in plain language — the score mechanics, the five categories and their weights, each evidence term, the signals reported alongside the score, and the verification proof — in all eight languages. Linked from the navigation, the footer, and every report, with each category deep-linking to its definition.
v1.8.0 — pending
A new look for the badge. The embeddable score badge — both the nameplate and the round seal — is redrawn on a polished metallic frame in the brand colors, with the score, band, and domain composited on top. The score number keeps its trust-band color so the result still reads at a glance.
v1.7.0 — pending
Report this score. Every report now carries a feedback control: pick a reason (scam, misleading content, wrong identity facts, score too high/low), add context, and it lands in a human review queue in the admin console. A confirmed report can trigger a re-scan or a correction — reports never change a score automatically, so scores stay reproducible from evidence.
v1.6.0 — pending
Sharper messaging & an owner invitation. The homepage now leads with what the product promises — "Trust you can check, not guess." — and says plainly that when there isn't enough evidence to score a site, we say so instead of inventing a number. The primary action is a free Quick Score, and a new "For site owners" section invites owners to verify their domain, embed their score badge, and back it with an on-chain attestation. Malware/phishing checks (URLhaus) also now run on every tier — see the model history for the collection change.
v1.5.0 — pending
Quick Score. The free scan is now presented as the Quick Score — the checks we can verify in seconds — with clearer messaging about what the Premium Advanced scan adds today (Google Web Risk threat intelligence, full per-signal evidence, fresh re-scans, exports) and what is coming to it next (reputation history, reviews & mentions from around the web, content evaluation). Reports that lack enough collectable evidence now say "Not enough evidence" instead of showing a misleading number.
v1.4.0 — 2026-06-27
Verification & documents. The on-chain score attestation is now offered only to a verified domain owner — with a "verified owner" indicator — and each domain has a single verified owner. When a domain changes hands, the new owner can transfer it to their account by proving they control its DNS. The signed PDF certificate and the printable report were tightened to fit a single page.
v1.3.0 — 2026-06-24
On-chain attestations. Every score can now be cryptographically attested on Base with the Ethereum Attestation Service, alongside a public domain directory and a verification "authority" hub where anyone can check a score independently — without trusting us.
v1.2.0 — 2026-06-23
Refreshed design & promotions. A new design-system look across the site and both consoles, a promo-code redemption system, the ™ mark on the wordmarks, and a richer contact form.
v1.1.0 — 2026-06-21
Accounts & billing. Metered plan entitlements with checkout and a usage dashboard, plus an email-confirmation step for new sign-ups.
v1.0.5 — 2026-06-14
Admin insights & reliability. New admin tracking — traffic, a conversion funnel, an activity feed, and an audit log — and self-healing database migrations so schema changes roll out without manual steps.
v1.0.4 — 2026-06-13
Analytics. First-party scan-traffic and marketing-source capture with an admin Traffic panel, and a daily data-retention sweep.
v1.0.3 — 2026-06-12
Transparency. A public Model-history page with an automatic version recorder, and admin controls for the premium-signal API keys.
v1.0.2 — 2026-06-11
One engine, redesigned badges. WebTrustScore and DomainValueScore now run on a single multi-tenant platform, and the embeddable badges were redesigned as a medallion seal.
v1.0.1 — 2026-06-10
Eight languages. Full interface translation into English, Spanish, Traditional Chinese, French, German, Arabic (right-to-left), Portuguese, and Japanese, with a flag language picker.
v1.0.0 — 2026-05-29
Public launch. The 0–1000 website trust score across five signals — security, ownership identity, reputation, transparency, and resilience — with free quick scores, premium detailed reports, signed attestations, embeddable badges, and accounts.