Monitoring
Watches what you run, around the clock.
Xovris checks whatever you run that the internet can reach – sites, services, domains and devices among it – from places around the world, and tells you only when something is really wrong.
XovrisOn its own
- A lit square is a feature people use today
- Newest: The free check reads cookies and trackers
Every feature in Monitoring, strongest first.
What fixes and prevents a problem comes first. Type to find one, or open a plan to see only its own.
Nothing here matches those words yet. Xovris notes the words, with no address and no cookie, so what is missing gets written.Ask Xovris instead
Lit: live and workingOpen: Xovris is still working on it
The free check reads cookies and trackers
The free website check names the cookies a page sets before anyone agrees and the trackers it loads, worst first among every other finding.
Live logs you can search in your own words
Open your logs, search every stored line in plain words - 'errors from checkout in the last hour' - and follow new lines as they arrive.
Log lines arrive within half a second
Following your logs live, each line arrives within half a second, and none is sampled away.
Know when logs stop arriving
Xovris notices when a source stops sending logs or a collector stops passing them on.
A quiet log source noticed fast
A source that usually sends on a steady beat is called quiet within two of its own intervals.
Ask 'what changed before yesterday's outage?'
Search your logs in plain words or in the query language, and read what changed just before a problem.
Search logs in your own words, query shown
Ask your logs a question in plain words and see the exact query Xovris ran to answer it.
Xovris reads everything you connect openly, never from a list of checks
Every record a connection returns is read whole: each measure against its own normal, each state against its last good, every change in the window, every source that disagrees with another about the same thing and every source that went quiet. Whatever is out of place is a finding of its own kind, on every plan, before any AI is asked - nothing has to be on a list first.
Logs open on their volume by level
The logs page opens on a strip of how many lines arrived at each level, errors and warnings coloured, above the live stream.
Speed promises written from your real traffic
Xovris writes each journey's objective from its real traffic, in plain words, with a speed promise you can publish.
Real-user Core Web Vitals, with the culprit named
Each real user's LCP, INP and CLS is kept with the element or interaction responsible.
Web vitals read with Google's own library
The browser SDK reads LCP, INP, CLS, FCP and TTFB through Google's own web-vitals attribution build.
A device that blocks ping is not called down
When a device ignores ping, Xovris asks its services instead, so it is never reported down just for blocking ICMP.
Every log step runs for real
Every step Xovris hands you from a log explanation is one it carries out itself on the machine.
Chart steps done by Xovris
When a chart's explanation points at a step, such as comparing with a committed baseline, Xovris does it for you.
Every prediction shows how sure it is
Each forecast shows how likely it is and its record so far, so you can judge it yourself.
OpenTelemetry exponential histograms read in full
Xovris reads OTLP exponential histograms with every bucket, so percentiles from your services are exact.
Log archives find their own bucket
If no bucket is named for a log archive, Xovris reads the buckets your credentials can see and uses the right one.
Checks set up from an address alone
Give Xovris an address with a port, a DNS name or a password inside it, and it works out the check, seals the secret and sets it up itself.
Every log explanation's step is a button
When Xovris explains a log line, each step it hands you is a button pressed right there, with Xovris doing the work.
A small slip in a check is corrected for you
A syntax slip in a check you wrote is corrected by the parser's own fix, with no AI involved, and the check runs.
Log fixes pressed where the line is
The fix for an explained log line is carried out from the line itself, not from a separate page.
Log forwarders set up with one press
On the log forwarders page, 'Let Xovris do it' sets the forwarder up for you, right where the line is.
Certificates renewed at the right moment
The certificate watch decides each name's renewal itself: nothing before its renewal point, and once past it with no successor, the renewal is made.
Proof that a prevented problem was real
For each problem Xovris prevents, it keeps the evidence that it really would have happened.
Status page updates drafted for you
Beside each open incident, your status page offers an update drafted in plain words from the incident's own record.
A vendor outage anywhere, your workspace checked
When a vendor you use breaks, Xovris checks your workspace and protects it within fifteen minutes.
Any vendor's status, not only the listed ones
A vendor that is not on Xovris's list has its status page found from your site's own records and read just like the others.
Every change watched closely for a day
For a day after a change, Xovris reads your checks' response and DNS times every five minutes and judges the change on them.
Certificate renewals raised once, with the full picture
When a certificate is due, Xovris raises it once with what will happen, when, how sure it is, what it is doing and how to undo it - and closes it the moment it is renewed.
Small drifts caught before they become outages
Xovris watches your own metrics for memory creeping up, retries rising and queues growing, and warns you while there is still time to act.
Your site checked from four countries every hour
Every hour, Xovris checks your websites and DNS from Germany, the United States, Singapore and Brazil, on several networks.
Every provider you use is watched
Xovris watches the status of every provider your workspace uses - not a fixed list - and tells you when one of them has trouble.
Busy days predicted from your own traffic
Xovris predicts your busy days from at least two weeks of your own traffic and your country's calendar, so it can prepare before the rush.
Disk and battery health read from each computer
Each computer reports its drive's failing sectors, its battery's wear and its temperatures, so Xovris can warn you before the hardware gives out.
The next device of a make connects at once, even when it is set up differently
Once Xovris has worked out how to read one device, the next one of that make is read in full straight away - including one whose owner set it up on different ports. Xovris recognises it by what the device itself answers, and anything that only looks similar is checked from the start instead.
Told when a backup fails, and when a scheduled one never starts
A backup run that fails is raised the moment its line arrives, each job shows how old its last run is, and a run delivered twice is counted once. When a scheduled backup never starts, Xovris raises it once – it checks every minute.
Domains shows DNS changes, names that stop resolving, and domain and certificate expiry in plain words
Domains shows when a DNS record changed, with the before, the after and the time. It shows when a name stops resolving, with the answer the resolver gave. You are told 60, 14 and 7 days before a domain’s registration ends, from the registry’s own record. Each certificate is judged by its own lifetime, so a short-lived one warns at its own renewal point. An expired certificate, a name it does not cover, a wrong hostname and an incomplete chain are each said in words.
Every area Xovris looks after is a lamp lit by a real check
Wherever the areas Xovris looks after are listed, each is now a lamp in the state Xovris’s last real check of its own domain left it, with its count said beside it – and a check is counted under one area only, so what the lamps say adds up to the record. Each kind of device opens on one real device drawn as a ring of what was read from its own answer: a reading the device did not give is grey, never green.
A failing website check says exactly what a browser got, and from where
When a check fails, Xovris says the status a browser received and the places that saw it – for example HTTP 503 from your checkout page where 200 was expected, seen from London and Frankfurt – never an internal code.
Search your logs in your own words, follow them live, and choose how long they are kept
On Logs you ask in plain words, such as errors from checkout in the last hour. Every answer cites the lines it rests on. You can follow lines as they arrive, with new patterns marked. Your plan’s window is stated. Anyone allowed to change settings can switch on a retention preset for an obligation such as PCI DSS, and each change is shown as a dry run first. Logs kept past the searchable window stay reachable. When a source goes quiet for three of its usual intervals, or a collector stops passing lines on, Xovris raises one incident saying so.
One box runs every free check on what you type, and each card lights with your own answer
On the free tools page, type an address, an email, an IP, a repository or a token once: every check that can read it runs at the same time, and each check’s card and its lamp on the board show your own answer the moment it lands. Until you type, each card shows the answer for Xovris’s own domain with the time it was read. Nothing on the page moves when a run ends, and one press puts everything back.
Check what only your own network can reach, from inside it
On Monitors, one button starts a private probe on a computer of yours that already runs the Xovris app, so a system reachable only inside your network is checked from there. Its token goes sealed to that computer once and never appears in an answer, report or log.
Tell Xovris that what a thing reads now is its normal
When something you watch has changed for good, say so and Xovris checks it again straight away. A steady check takes that value, and a range starts again from it and is worked out again from what that thing reports.
On the status page a check to improve carries an amber mark, never the green one
Every row of Xovris’s own check carried the same green mark – the two rows that say Improve among them. A row’s mark now follows its state: the logo’s green only when both readings are fine, amber when one says improve, red when one says fix, the same colour as the word beside it.
The ring shows everything you run whole, and names what needs you under it
The ring of everything you run on the home page now fits inside the window at every size. Its middle counts the coloured arcs and names each one, and the readings behind them are listed under the ring in plain words, each one press from its own page.
The ring on the home page and the record under it say one thing
The ring now counts exactly what the record of the check under it counts, in the same words. A check to improve turns its area’s lamp amber, the middle says how many to improve and how many are fine, and whose reading it is and when. What your own browser reads about this device and this connection is said apart, under the ring, by name – never counted under another name or time.
Told before a disk fills
From what a device reports, Xovris works out the day its disk will be full. A fast fill this week is not hidden by ninety quiet days before it. That day is shown beside the other things about to fail, and the fix is scheduled before it. On a computer running the Xovris app, a system disk nearly full, a backup switched off and a backup gone stale are each said in plain words.
The trust page opens on its live status, the free tools always answer, and what Xovris does is a board of lamps
The trust page kept its one real reading – Xovris’s own checks from outside – inside the first of fourteen closed sections. It now opens on that reading as a ring, one arc a check, and its facts stand open. On the free tools page an address and Enter always answer: a press is never lost, the button can be pressed again while the checks run, and a check that has not answered in ten seconds is said so. A tool’s own page shows its recorded answer for Xovris’s own domain before you type. What Xovris does today is a board with one lamp a line and cards in full rows, never a wall of links, and every feature’s nine areas are a board that opens in place. The developers page sends a real request from its first screen, and the network map stands beside its headline with no line drawn through a name.
The free website check now reads security headers, email, DNS, uptime worldwide, links, cookies and more
You can now type an address and the free check answers in seconds. It grades the security headers and says when the certificate and the domain registration expire. It checks email delivery (SPF, DMARC and DKIM) and DNS health, including a name still pointing at a hosting service that no longer exists. It checks whether the site answers from every inhabited continent. It also reads the page itself: accessibility basics, up to 25 broken links, search basics, the cookies set before anyone agrees, the trackers it loads and what the site runs on.
Find the waste in your object storage
Xovris reads your storage bucket’s own inventory and shows where the money goes on tiny files, duplicates and data nobody reads, with what each change would save.
Protect every member of your household online
A parent adds each member of the household once, and Xovris’s protective web filtering looks after them on the devices they use.
One agent brings in the logs of every computer, cloud and office suite you run
You can now install one Xovris agent on Windows, macOS or Linux and the computer’s own event log arrives in Xovris. See who changed what in AWS, Azure and Google Cloud, and the sign-ins, sharing and mail changes of Microsoft 365 and Google Workspace, beside everything else you watch.
A lookalike site's abuse report is filed by Xovris the moment it can be, never left to you
When Xovris finds a site imitating yours and the abuse desk is not taking reports at that moment, Xovris keeps the report and files it itself the moment the desk opens - it never hands the report back to you. It keeps the report open, checks every pass whether Netcraft's reporting or the registrar's abuse mailbox can now be reached, files it the moment one can, reads each one back and tells you. The written report stays beside it if you would rather send it yourself.
Watch hosts, virtual machines, containers, Kubernetes and cloud services
Point the OpenTelemetry collector you already run, or Xovris’s agent, at Xovris and your hosts, virtual machines, containers, Kubernetes and cloud services are watched beside your sites and apps.
Keep harmful websites off your home or office network, with a way to ask for one back
Xovris can filter the web for everyone on your network through your own gateway: it judges each site as it is visited, applies the policy you choose, checks that your gateway really took it, shows a clear page when something is blocked, and lets anyone ask for a site to be opened with one press.
Glyph shows how xovris.ai is doing, in the header
On a wide screen the status in the header of xovris.ai now carries Glyph – proud while every recorded check passes, concerned when one does not and curious during maintenance – read from the same live status as the status page.
Glyph offers to check your last address again – only if you said yes
When a check finishes on xovris.ai, Glyph asks before it remembers anything. Say yes and only the address, its result and the date are kept, on your device alone; on your next visit Glyph welcomes you back with that check and offers to run it again, and one press forgets it. Say no and nothing is kept.
Ask Xovris to check anything you run, right in the chat
The check box in the chat on xovris.ai now takes a website or domain, an email address, an app’s API, an IP address or a code repository and checks each as what it is, line by line, from your browser. Type a device – this one, your router or a printer – and it offers the check of this device or this connection now, and says how the rest are looked after. The chat’s suggestions now give every area its turn and lead with the area of the page you are on.
Every technical word explained on the page that uses it
When a page on xovris.ai uses a technical word – the address book of the internet, a program-to-program connection, a message sent to your own system – it now says what that word means in plain English at the end of the same page, so nobody has to leave the page to look it up. The list is made from what each page actually says, so a word added tomorrow is explained the day it appears.
See your home or office network as a map you can share
xovris.ai now draws what the Xovris app finds on your network as a map: the router in the middle, every device around it named from what it answers, and anything that needs you ringed and listed first. Share it as a link – the map travels in the part of the link a browser never sends to any server, with addresses left out unless you include them – or download it as a picture. In the Xovris app, press See this as a map you can share after looking at the other devices.
One tap to be told when something is back
When xovris.ai finds a service or any website down for everyone, one tap beside the answer asks your browser once for notifications and keeps checking from your browser. The moment it answers again you are told on the page with the time, by a notification, in the tab’s title and by Glyph in the tab’s icon – even while you work in another tab. It says plainly that this works while the tab stays open.
See the whole of Xovris in one place
The features page now opens onto every part of Xovris, grouped by what it does for you – setting up your tools, fixing problems with proof, finding the cause, looking after what you connect, keeping your family safe, working with your people, saving money and building on Xovris. Each part lists the features that do its work, each with what it does for you.
See what an outage somewhere else would do to you, before it happens
Xovris keeps watch on the providers and regions that many services lean on and follows outages around the world as they happen. Ask what would happen to you if one of them failed and Xovris works it out on a copy of your own setup – never on the real thing – and warns you early when one of them starts to wobble.
xovris.ai shows Xovris at work – try it before you sign up
On xovris.ai you can now watch Xovris put right something a visitor breaks, see a device join a network and get looked after, talk to Xovris by voice, explore everything it looks after, read the page written for your household or your kind of business, work out your price with a calculator and read in plain words what happens to your data. The site adapts to what you tell it you need, and what it says Xovris does is shown working.
Lost on a page that does not exist? Catch the broken link
An address that leads nowhere now shows the pages closest to it straight away, and a small game beside them: Xovris chases the very link you followed – point or use the arrow keys – and, caught, it turns into those pages. It works by keyboard and screen reader, and with motion off nothing moves.
Share a check’s result as a card
Every finished check on xovris.ai can be shared as a picture: Glyph with the address, the counted result, the date and a link to check it again – nothing else. On a phone it goes straight to your share sheet; elsewhere the picture is shown to download and the link is copied.
Every device you connect named by its maker, model and exact version – cameras, printers and power included
Xovris now recognises each device from what the device itself answers. Over SNMP that is more than fifty kinds of network gear, storage, printers, cameras and power – Cisco, Juniper, Fortinet, Palo Alto Networks, Aruba, Ubiquiti, MikroTik, DrayTek, Synology, QNAP, TrueNAS, HP, Canon, Brother, Xerox, Ricoh, Hikvision, Dahua, Axis, APC and more – and from the system itself on Windows, macOS and Linux computers, Android phones (by their security patch level, never their release alone) and iPhone, iPad, Apple Watch, Apple TV and Apple Vision Pro through device management. Anything no rule knows yet is looked up in the official product dictionary that security notices use, refreshed every day. Each device recognised is checked for known security holes, worst first, with the upgrade that closes the most.
Your DNS backups are proved every day by actually restoring them
Each day Xovris restores your zone’s saved records into a separate copy, looks every record up and compares it with the original, down to long DKIM keys. Each zone shows whether its backup is proven, partial (naming the record types it cannot restore yet) or failed, record by record – so you know a backup works before the day you need it.
Domain renewals warned about 60 days ahead, and only while auto-renew is really off
Xovris reads whether a domain renews itself from the registrar itself, through your own connection, and warns you 60, 30, 14, 7 and 1 days before it lapses – earlier than the registry’s own reminders. If auto-renew is on, you get a quiet line in your digest instead of an alarm, and if the renewal still has not happened when it should have, Xovris alerts you again.
Drop a link onto Xovris the helper and it checks it at once
Drag a web address or an email address onto Xovris the helper on xovris.ai – or paste one onto it on a phone – and the free check starts straight away in its chat, each result shown as it comes back.
Every device you connect, checked for known security holes, worst first
Firewalls, network storage, switches and computers are each recognised by their exact version and checked against the makers’ own security notices and the lists of holes attackers are using now – worst first, with how exposed each device is, how much your business depends on it, whether its maker still supports it, and the one upgrade that closes the most. Xovris carries out each upgrade in order, within your policy, and calls it done only when a fresh read shows the new version.
Any sentence you select, explained in plain words
You can now select any sentence on xovris.ai and a small Explain this appears beside it (below it on a phone). Press it – or Alt+X on a keyboard – and the sentence goes to the helper as your own question, answered in plain words.
Free certificate and domain expiry alerts for anyone, with no account
Give Xovris an email address and the sites you care about, confirm from the email, and Xovris checks each one every day: the certificate it serves and the domain’s registration, read from the registry itself. When either is close to lapsing you get one email per site, never twice for the same warning, with the choice to let Xovris fix it. Where a registry does not say whether a domain renews by itself, Xovris tells you instead of assuming it does.
Pick Glyph up and drop it in any browser – and feel it on a phone
On xovris.ai you can pick Glyph up, carry it and let it go: it swings from where you hold it, falls, lands and stays where it landed on the next page. It now behaves the same in Chrome, Firefox and Safari’s engine, a fast flick no longer throws it out of sight, and on a phone a short tap of vibration tells you when it is picked up and when it lands – never with motion turned off.
Glyph takes the shape of what each page is about, and moves in only one tab
Arriving on a page about websites, apps, domains, email, security, networks, devices, the cloud or integrations, Glyph’s X becomes that area’s own shape for a moment – a window, code brackets, a globe, an envelope, a shield, wifi, a phone, a cloud or a plug – and returns to the exact logo. With xovris.ai open in several tabs, only the tab you are using has a moving Glyph; the others rest until you come back.
Glyph’s sounds, only if you want them
Glyph can now make small sounds on real moments, such as a check finishing. They are off unless you turn them on with the Sounds setting at the foot of every page, the choice is kept only on your device, and they never play with motion turned off or in a tab you are not looking at.
Your browser tab shows the check as it runs
While a check runs on xovris.ai, the little icon in your browser tab becomes Glyph acting it out – a ring that fills as each answer lands, then worried or pleased by the real result – so you can work in another tab and still see how it is going. It listens while the helper waits for you and thinks while an answer is on its way.
Ask the helper on xovris.ai about anything you run
The helper on xovris.ai now answers for every area Xovris looks after – websites, apps, domains, email, security, networks, devices, the cloud and code, and anything else you connect – and can take you straight to that area’s free check. It starts with what the page you are on is about, says plainly when the site does not say, and never pushes you.
Pages stay smooth with Glyph moving, in more browsers and on slower phones
The soft lights behind the top of each page are now drawn without a blur, looking the same, so Firefox draws the home page at its full frame rate with Glyph moving; a device that still cannot keep the lights moving smoothly holds them still. Glyph’s mood follows your own time of day – sleepy late at night, bright in the morning. The search also closes in the same instant you press Escape, even on slower phones.
The status page lists every time Xovris fixed itself
xovris.ai/status now shows each time Xovris healed its own service: what stopped working, what Xovris did, how it was undone and whether it checked out. Xovris only fixes itself in a few pre-approved ways that undo themselves; anything bigger waits for a person, and some changes – to its own code, its own permissions or how you are billed – it never makes to itself at all.
Xovris the helper naps while you are away and wakes when you return
Switch to another tab and, once nothing needs it, Xovris the helper settles down for a nap; come back and it stretches awake where you left it. With motion off it simply waits, still.
Ask Xovris out loud
The helper on xovris.ai now has a microphone button beside Send. Press it and speak: your words appear in the box as you talk and your question goes when you stop. If the microphone is blocked, missing or hears nothing, it says what to do next – and where a browser cannot turn speech into text, it says so and suggests your phone keyboard’s own microphone. Late at night Xovris yawns now and then, and a first visit can start the one-minute tour straight from its hello.
Clearer words across xovris.ai
Forms now say each field’s own problem beside it, every free check gives its verdict in words – “Not listed: all 3 addresses were clear on all 4 lists asked” – the coverage page counts only what you can use today, and the changelog, AI, trust and results pages say what you get and the guarantee you can hold Xovris to.
Connect your app your way: Xovris does it, the wizard guides you, or you do it by hand
The connect section on xovris.ai now asks how you would like to connect. Let Xovris do it for me opens the pull request Xovris makes for you, with Vercel beside it. Run it from my terminal opens the one-command wizard; and I’ll do it myself links each stack’s guide to every file and variable the wizard would change. Every way ends in the same verified connection, and the site remembers your choice.
A ready dashboard for everything you connect, the moment its first data arrives
Connect a website, an app, a log source or a server and its dashboard is made for you as soon as its first data lands – each chart with one plain line saying what it shows. A chart with nothing to draw yet says what it will show and why it is waiting, instead of sitting empty.
Every page about Xovris starts on one thing it does for you – some shown live in your own browser
About, Careers, Contact, Cookies, the changelog, Credits, Privacy and Terms each now open on one fact with the Xovris mark beside it, and the mark folds into its tick the moment that fact lands. Cookies counts the cookies your browser holds for this site and Privacy counts the other sites this page asked for anything – both measured in your browser as you read, never claimed for you. The docs header now has Sign in and Start free, and every plan card says plainly which figures it carries.
Three new free checks: an API, an IP address and a code repository
The free tools now cover your apps, networks and code as well as your website, domain and email. The API health check calls any API address and times its answer, the IP address checker shows who announces an address and whether its route is signed, and the repository health check reads a public GitHub or GitLab project’s branch rules, security policy and OpenSSF Scorecard – all in your browser, with nothing sent to Xovris.
Four more free checks: your browser, your connection, who hosts a website and your dependencies
The free tools now cover devices, cloud and code. The browser and system check says whether your browser is on its maker’s current release and whether Windows still gets security updates; the dependency check reads a package.json, package-lock.json, requirements.txt or go.mod and shows every package with a known vulnerability, how serious it is and the version that fixes it; the connection check shows your connection as the internet sees it; and Who hosts this website? names the host and DNS provider of any site. Each runs in your browser, with nothing sent to Xovris.
Three more free checks: read a login token, check passkeys and find your nearest cloud region
The JWT decoder reads a token’s claims and expiry in your browser and flags the faults that matter; the passkey check asks your browser what it can do with passkeys on this device; and the cloud region latency test times every AWS region from where you are and lists them nearest first. Nothing is sent to Xovris.
Xovris the helper reacts to each line of a check as it lands
While a check runs on xovris.ai, Xovris reacts to every result the moment it comes back – a nod when a line passes, a moment’s thought for advice, a wince when it finds a fault, a look about when a line could not be read – and goes back to reading until the last line is in. With motion off it holds still.
Xovris the helper never rests on what you are reading – on a phone too
Before it settles, Xovris now measures every link, button and line of text on the screen – its own put-away button included – and moves on when the page changes under it, for example when a list finishes loading. Where a phone screen has no clear space at all, it waits in the logo at the top until you scroll somewhere it can stand.
The helper on xovris.ai comes back on its own if it ever fails
If Xovris the helper or its chat ever hits an error on xovris.ai, it now restarts itself a moment later instead of disappearing until you reload, and the error reaches Xovris’s own team the same day. Before any new version of the site goes live, every button on the helper and the chat is pressed in a real browser on a phone-sized and a desktop-sized screen, and the release stops if anything breaks.
The helper keeps answering even when part of it is down
If anything behind Xovris the helper fails, it carries on answering under the same rules on what may be said and what is kept – so a question still gets its answer.
The helper’s chat opens and minimises the same way on a phone and a desktop
On xovris.ai the chat with Xovris now grows out of it on a phone just as on a desktop, Xovris never sits over the chat’s own words while it opens or when minimised, and minimising does the same thing on every screen. Before each release this is checked frame by frame in a real browser at phone and desktop sizes.
The free check on xovris.ai now takes anything you run
You can now type a website, an email address, an IP address, an API or a code repository into the box on the first page and Xovris works out which it is and runs that kind’s own checks in your browser: where an address’s mail goes and every protection on it, who announces an IP and whether its route is signed, how fast an API answers and who may call it, and a repository’s branch rules, security policy and OpenSSF Scorecard. A website’s certificate and its page from three other continents are answered too, so no line waits for later. An email address never leaves the page – only its domain is checked.
Play with Xovris: it chases your mouse, and you can catch it
Press Ctrl or Cmd K and choose Play with Xovris: the helper runs after your mouse, always beside the pointer so it never covers what you point at, hops aside if you swoop at it, and celebrates when you catch it. Press Escape, scroll or carry on with the page and it walks back to its place. On a touch screen it tells you to pick it up and drop it instead; with motion turned off it does not play, and says so.
See everything you can use on xovris.ai today
The coverage page now lists every free check you can run right now – sixteen of them, and a live is-it-down page for each popular service – each with what it tells you. The count comes from the same lists that build those pages, so it can never say less, or more, than is really there.
Pick Xovris up, put it anywhere, and drop a link on it to check it
On xovris.ai, Xovris the helper can be picked up and carried: it dangles from your pointer, falls where you let go, lands on the card below and stays there, even on the next page. Drop or paste a link, a domain or an email address onto it and the check starts at once. Its conversation opens out of it, stays beside the page instead of covering it, and covers websites, apps, domains, email, security, networks and devices alike.
Xovris takes the shape of the page it arrives on
When Xovris the helper arrives on a page, its X briefly becomes what the page is about – a shield on security, an envelope on email, a globe on domains, a plug on integrations – then snaps back to the logo. It happens once a page and never with motion turned off.
A check of anything in your own words – continuously, hourly or daily
You can now say what to watch – “check my orders API every 15 minutes, expect 201 within 2 seconds” or a sign-in journey, each step named – and Xovris says back what it understood before it starts. Choose continuously, every minute, 5 or 15 minutes, hourly or daily; a key the API needs stays in your secrets, never in the check.
Know when an old DNS record lets a stranger take over part of your domain
Xovris now looks at every address record in your domain every five minutes for ones that still point at a service you no longer use – an old Azure app, an S3 bucket, a Heroku or GitHub Pages site, a domain nobody renewed, a cloud address you gave back. When anyone could claim it and serve pages as your name, you are told straight away with the proof and exactly which record to remove.
Xovris on your computer: it finds which part of your connection is at fault, and fixes it, asking only for what needs your hands
You can now install Xovris on a Windows, Mac or Linux computer and allow, in plain words, what it may check. It tests your connection layer by layer – the Wi-Fi or cable, the router, your provider’s line and network – and names the one at fault with the measurements that prove it, and a letter for your provider when it is theirs. It finds a firewall left off, a full disk or a crashing app and fixes what you allowed, testing before and after. When a cable needs your hands, it asks for that one act and checks it worked. Every check is in a log you can read, and one press stops it all.
Xovris on your computer finds the other devices on your network and says what is wrong with each
Press Look at the other devices in the Xovris app and it reads every device on the network your computer is on, names each from what it answers – the router, a printer, an iPhone or iPad, a storage box, a smart-home hub such as Home Assistant or Philips Hue, another computer – and says what is wrong in plain words: a printer out of paper, jammed or low on toner (read from the printer itself), a router answering slowly, a storage box whose file sharing stopped, the smart devices a Home Assistant hub cannot reach. A device it cannot name yet is listed with what it answered and the one thing Xovris needs to check it properly. It only reads, and only on the network you are on – and the map it draws is yours to share.
Domains, DNS and registration watched – and every change explained
Xovris reads each domain’s registration from its registry – expiry, renewal, transfer lock and registrar – and its DNS, and tells you in plain words when anything changes or is about to lapse.
Is it down for everyone? Asked from every continent at once
The website-down checker now asks your address from one place on each continent at the same moment and shows each one on a map – its city, its answer and how long it took – beside your own browser’s check. It tells you whether the site is up for everyone, down for everyone, up in some places and blocked or failing in others, or fine everywhere but your connection, always with the reason. The DNS change checker shows the same places looking your name up through their own local DNS.
From any log line, open its trace, the release that was live, its incidents and what changed just before
Press any line and Xovris shows everything around it: the whole request it belonged to across your services, the release that was running, every incident it is part of, the thing it came from with that thing’s history, and every change from every connected source in the minutes before – with no set-up.
Four more free checks: blocklists, security headers, Cyber Essentials readiness and leaked passwords
See by name which email blocklists list your address or domain, with each list’s removal page; your site’s security-headers grade with the exact lines for your server; your Cyber Essentials readiness from ten questions, each tied to the requirement it comes from and read across ISO 27001, SOC 2 and NIS2 – readiness, never a certificate; and whether a password is in a known breach, without it ever leaving your device. Every report run on an address offers to watch it free.
Watch up to 500 certificates free, warned at the right moment for each
Since Let’s Encrypt stopped its reminder emails, nobody tells you when a certificate is about to lapse. On the Free plan Xovris watches up to 500 names, warns you when each certificate should be renewed by its own lifetime – so shorter certificates are not left until a fixed 30 days – and tells you when a certificate is issued for your names that nobody ordered.
The free website check stays free for people and closed to robots
Check three different websites a day without an account; checking one of today’s again is always free, and a site you prove is yours with one DNS record never counts. People never see a puzzle: a quick background check runs in your browser, a person check appears only from your second check, and no website can be flooded through the checker – at most thirty different people can check one site in ten minutes. Signed in, the allowance does not apply at all.
Your free checks are your own, and nobody can use them against a site
Each visitor keeps their own free checks even on a shared office or home network. An address with a port the check does not test is told which ports it does and how to write the address instead. And a website can never be flooded through Xovris’s free check: when many people check the same site at once, they share one result and the next person is told the site is being paced and when to try again.
Every free check now does the whole job, in your own browser
Under each free check is the full report: what 22 public resolvers answer for your name and whether a change is live everywhere; every record with its DNSSEC chain link by link; whether Gmail, Yahoo and Outlook will accept your mail; your domain’s expiry and the locks that stop a hijack; whether a message really came from its sender; whether a site is down or it is you; subdomains someone else could claim; and whether your certificate renewals survive the shorter limits of 2027 and 2029. Nothing goes through Xovris unless you ask its probe.
Is GitHub, Slack or Gmail down? Checked from six continents when you ask
Pick one of 23 popular services and Xovris checks it from six continents the moment the page opens, asks it from your own browser to tell you whether the problem is yours, and shows what the service’s own status page says beside it – saying so when the two disagree. Every answer links to its raw record; nothing is a count of user reports.
Watch the journeys your visitors actually take
Ask Xovris which journeys to watch and it reads the visits your own site logs already record: the path most people take – home, a product, the basket – with how many visits took it, ready to add as a check in one press. Nothing is saved until you say yes, and the visitor's session itself is never stored.
Journeys ready for sites like yours
When your shop or blog runs on a platform others use – Shopify, WordPress – Xovris offers the journeys proved working on other sites of that platform, such as a checkout, with how many sites watch it and none of their details.
A live badge that proves your site is safe, and stops when you say
Put a Xovris badge on your own pages: it reads verified only while your site passes Xovris’s hourly check from outside – HTTPS answers, the certificate is valid and DNS resolves – says why when it does not, and reads not verified the moment you revoke it.
Watch log lines arrive, ask your records directly and explain any error
Logs now streams new lines as they are stored and marks the first of a kind never seen before, picking up from the exact line after a break. Ask your records in a query that is checked as you type, save it for everyone or just you, and paste any error to read what it means and what to do – keeping the pasted lines only if you choose.
Logs that never hide an error, answer questions exactly, and cost less without losing evidence
When a sender’s format changes and lines stop being read, you are told within minutes how many lines from which source, and none is lost: each is kept as searchable text or set aside with its reason, and a line that says it is an error is never filed as routine. Plain questions over your logs are answered exactly, each sentence pointing at its lines. And routine noise is trimmed only where no investigation would lose its evidence – your own rules always win.
When a system changes how it writes its logs, Xovris reads the new shape and no error is hidden
A software update often changes the way a server, router or app writes its log lines. Xovris notices the moment its lines stop being understood, works out the new shape from the lines themselves, checks that nothing it already knew is read worse, and carries on – so an error is never hidden because a format changed.
Logs you can trust: private by default, gaps named, attacks recognised
Web server and CDN logs arrive with visitors’ addresses shortened and secrets masked before they are stored; a source that goes quiet or loses lines is named with how long; and sign-ins, account changes and cleared event logs are recognised by the same open detection rules security teams use.
Every log, check and change in one place, in order, with secrets masked
Logs from your apps, the result of every check and every change Xovris makes land in one searchable place, kept in the order they really happened across machines, with passwords and keys masked before anything is stored – and every kind of line counted, so a new one stands out.
Software Xovris has never seen is watched properly the day you connect it
Connect a database or service Xovris has no ready-made integration for and it works out from the service’s own description what shows it is healthy, tells you in plain words what it will watch, checks it every five minutes, and when something goes wrong offers the fix that service itself provides – already tested so it can be undone.
The free scam checker and your command centre now judge every message the same way
The checker on xovris.ai and Xovris’s own “is this a scam?” share one set of rules – a request to send back a code, a small fee to release a parcel, a “new number” message, a brand writing from a free mailbox – and if you already clicked, typed a password or paid, the advice now fits what you did.
Is this a scam? Give it to Xovris and find out, with every reason
Hand Xovris a message, a link or a website and Xovris tells you whether it is a scam, looks suspicious or shows no signs – never “safe” – with every reason it saw: pressure to act fast, a request for a code or a gift card, a link that hides where it goes or borrows a company’s name, a website registered days ago. Then exactly what to do now.
Change how often a check runs and see what it changes first
Before you save a new schedule for a check, Xovris says back the runs a month now and after, how soon a problem would be seen on average, and the fixes it handled for you from your own last 90 days – and nothing is estimated when there is not enough history yet.
Grade your security headers from your pipeline, free
The security-headers checker now answers as an API on every plan, including Free: your site’s grade, every test behind it and the exact lines for your server, as JSON. A ready workflow grades your live site after each deploy and fails the build when a header goes missing, and moving from the securityheaders.com API takes three small changes.
Cyber Essentials Plus, DORA and PCI DSS read from the same evidence
Your Cyber Essentials readiness is now also read against the five tests a Cyber Essentials Plus assessor runs, DORA’s ICT risk-management standards and PCI DSS v4.0.1 – each control counted only where your result fully stands for it, with what is still yours to do (the assessor’s own scan, PCI DSS’s stricter password rule, DORA’s incident reporting) named beside it. The free checker on xovris.ai shows the same.
Anything that stops answering is found from outside and fixed
Pages, APIs, sign-in journeys and cron jobs are checked from Xovris’s probes on a schedule; a job that stops checking in is noticed by the silence, and every result is kept with where and when it was measured.
Tell Xovris what to watch, in your own words
You can now say what you want watched – “tell me if the card machine goes offline”, “warn me if anyone forwards company email outside”, “tell me when the supplier’s widget is back in stock” – and Xovris shows what it will watch, where it gets the facts, what it needs from you and the alert exactly as it will read. Send that alert as a test to your own email or Slack, then switch it on. Certificates, domains, pages, DNS, email protection, devices, sign-ins, payments, cloud changes, backups, jobs that should check in, your suppliers’ status pages and dates all work the same way.
A free speed test with real visitors’ figures
You can now type a page’s address to see how fast it was for real Chrome visitors over the last 28 days – LCP, INP and CLS, each rated by Google’s thresholds and explained in plain words, with the share of visits that were good – beside a lab run now and what would make it faster, biggest saving first.
See the weekly letter you will get, in your own voice
Each solutions page now shows the plain-language letter Xovris writes every week – for a person, a household or a business – on a clearly labelled example week: what it watched, found, fixed and proved, what it prevented, what it cost, and the one decision waiting for you.
Know when a certificate is issued for your domain
Xovris watches the public Certificate Transparency logs and tells you when a certificate is issued for one of your domains – including one you did not ask for.
The free website check shows what the outside world sees
You can now type an address on xovris.ai and Xovris checks it from its own probes outside your network – the certificate, the page and DNS – beside what your browser can see, and says plainly what it could not confirm.
An outage and a shaky connection told apart at home, with your home devices as sources
From the home’s own equipment Xovris tells a dead line, a lossy line, a lossy router and weak Wi-Fi apart and names each to its layer. A device that stops answering is raised once, on the first failure after it last answered. Home Assistant, UniFi and a home router’s log can each be connected if you want them: a device gone offline with what it last read, a threat detected and blocked, the internet connection down.
Glyph seals a proved fix with the Proof mark
When the recorded investigation on xovris.ai reaches the moment its fix passes the after-test, Glyph drops onto the proof, stamps the Proof mark and hops off, leaving the seal beside what was proved. With motion turned off the seal simply appears.
Your setup judged against Cyber Essentials, with the three fixes worth most
Xovris judges a business’s setup against Cyber Essentials v3.3 from what it reads itself. That covers firewalls, secure configuration, updates, access control and malware protection, for each device, account and service. A cloud sign-in without a second factor fails, and so does a high-risk update more than 14 days overdue. The monthly security score counts only what was measured. It names the three fixes that would raise it most, each with the points it adds and the button that makes it.
Spend finds waste and tight limits in your own usage, and judges each change once it is made
A database connection pool that ran full while requests waited is named with the limit to raise it to, and one with headroom is left alone. A host that never needed its size over a week, and graphics processors that sat idle, are named – a busy host, or one seen for only a day, never is. Each comes as the change itself with what was measured, and once the change is made the same measure says whether it got better, made no real difference or got worse.
Know when a build gets bigger, which package did it, and that the fix worked
Send each build’s size report from your pipeline and Xovris shows exactly what grew, file by file and package by package. When it grows past your limit you get one alert naming the change and the package, with the removal ready – and the alert clears only when a later build is really that much smaller.
Put the free website check on your own site with one line
Hosts, agencies and anyone with a website can paste one script line and the free check appears there under their own name, sized to its answers. Add a partner key and the sign-ups it brings are counted as yours. It sets no cookie and reads nothing from your page.
GPUs and AI models watched: errors alerted, idle time priced against your bill, slow models caught
Point the Xovris agent at your GPUs and the models you serve. A GPU hardware error reaches you once, in plain words with the first thing to do, and its all-clear arrives once when the GPU reports healthy again; a GPU that goes quiet stays open rather than quietly clearing. Each model shows how long requests take – for half, most and nearly all of them – how soon the first word arrives, and you are told when a model slows down. Give Xovris the line from your cloud bill and it tells you what idle GPUs cost you, but only when its own count of GPU-hours matches the bill.
What insurers ask for, beside Cyber Essentials
The Cyber Essentials readiness check now also reads the twelve controls cyber insurers ask about – multi-factor sign-in, backups, endpoint protection and the rest – from the same answers: met, partly met with what insurers also ask for, not met with the requirement that fails it, or not reached by the questions. Open any control from its own row to see what it reads. In the command centre the same twelve come from your real computers and accounts. Readiness, never a certificate.
From any log line to what changed before it, and its incident
You can now open any line to see the lines around it, its trace, the incident it belongs to and the change that came just before it – from deploys, settings and every change Xovris made – and when a log format changes, Xovris keeps reading it instead of losing the lines.
Say what to keep in your logs, and see the effect before it applies
On the Logs page, say in plain words what to keep, drop, mask, sample or send elsewhere – “drop health checks, keep errors from checkout, mask card numbers” – and see each part checked and what it would do to yesterday’s lines before anything changes. A rule that would remove a line an incident was proved with is refused, naming the incident. Once applied, each rule shows what it actually did against its preview, and every earlier version is kept.
Count anything in your logs, even lines you do not keep
You can now say what to count – “failed payments per minute by service” – or start one of the suggestions drawn from what past incidents relied on. Every line that arrives is counted, including lines your rules then drop, so the chart and its alert stay true while you store less. Set an alert when the count rises above or falls below a figure.
One readiness check, read across ISO 27001, SOC 2 and NIS2
The same Cyber Essentials results from your real computers and accounts are read across ISO/IEC 27001 Annex A, the SOC 2 common criteria and the NIS2 security measures – each control shown as met, not met with the requirements behind it, or not read yet – and each framework names what live evidence cannot show, such as training, audits and incident reporting. It shows where the work starts; it is never a certificate.
See every script on your checkout, and what changed
You can now run one bookmark on your real payment page and Xovris lists every script it loads – including ones added while the page runs, the way card skimmers arrive – where each comes from, whether its content is pinned, and what is new or changed since your last check. The list never leaves your browser.
A release that makes your app crash, caught from its crash rate and its store reviews – and every reviewer answered once it is fixed
Connect your app on the App Store or Google Play (and Sentry if you use it). Xovris reads the reviews and each release’s crash rate, and when a release crashes far more than the one before, or several reviews report the same problem, you get one incident – the crash rate, the reviews that report it and the error in your code they describe, side by side. When a newer release is healthy, Xovris drafts a reply to each reviewer; nothing is posted until you approve it, and it is never posted twice. With Business.
Your email addresses, domain and brand watched for breaches and lookalikes, and the response done for you
You can now add your email address, your website’s domain or your brand. Xovris tells you when an address you have proved is yours turns up in a data breach, and when someone registers a name that imitates your domain or brand – a letter swapped, a look-alike character, your name with “login” added – judged by when it was registered, whether it handles email, whether its page asks for a password and whether certificates were already issued for it. With your consent it ends every session and asks for a new password in your Google Workspace, Microsoft 365 or Okta, and reports a lookalike to its registrar and to the blocklists browsers and mail filters use. It tells you when a lookalike is taken down, and never raises a name you registered yourself. Your own addresses from Personal; domains and brand with Business Protect.
“Is this a scam?” answered with the reasons – and what to do next if you already clicked, typed a password or paid
You can now paste a message or a link. Xovris tells you whether it is a scam, suspicious or shows no signs of one – never “safe”, because no check can prove that – with every reason it found: a link to an address on a known-bad list, a domain registered days ago, a look-alike of a service you really use, pressure, a request for a password or a payment that cannot be reversed. If you already clicked, typed a password or paid, the next steps fit what you did. Only the verdict is kept – never the message. With Personal Protect; a few checks a day free on xovris.ai.
Your website’s quality checked in a real browser, then every day
Ask once and Xovris checks your site the way a visitor’s browser sees it – accessibility, whether tracking waits for consent, broken links, what search engines see and speed – keeps the report, alerts you once for each problem, and checks again every day.
The software on your computers checked every day against the vulnerabilities attackers are using now – and the one upgrade that closes them, done for you
The Xovris app lists the software installed on each computer and Xovris checks it every day against public vulnerability records, putting first what attackers are using right now and what is likely to be attacked soon. For each package you get one upgrade that closes everything fixable in it; press Fix and, with your consent on that computer, Xovris runs the upgrade and only calls it done when the computer’s next software list shows the new version. With the Xovris app on Personal.
Your existing browser tests become monitors
An existing Playwright test runs unchanged as a check from Xovris’s probes, so the journey you already test is the journey Xovris watches.
See every collector you run and fix one that stopped
Point your OpenTelemetry collectors at Xovris and see each one’s version, health, the configuration it runs and how much it sends. Send a configuration to all of them and see each confirm it; when one’s lines stop leaving, Xovris names the line that changed, counts what was lost and puts back the configuration that last worked.
Your Cyber Essentials readiness, read from your real computers and accounts – with the two automatic fails raised the moment they appear
Xovris reads what it already watches – your computers through the Xovris app, your people’s accounts in Google Workspace, Microsoft 365, Okta or GitHub, and the software updates still waiting – and shows each Cyber Essentials requirement as met, not met or not read yet, never assumed. A cloud account without two-step sign-in, or a critical update left longer than 14 days, would fail the assessment outright. Xovris raises each as an incident with what to do and closes it when it passes. With Business Protect.
The other areas
Words on this page, in plain English
Each technical word this page uses, said simply.
- Developer kit (SDK)
- A ready-made library a developer adds to an app so it can talk to a service.
- The open standard for app data (OpenTelemetry)
- A free, widely used standard for collecting an app’s errors, timings and records and sending them to any monitoring service.
- The open format for app data (OTLP)
- The way apps send their errors, timings and records in the open standard most monitoring services accept.
- Domain Name System (DNS)
- The internet’s address book: it turns a name you type into the number of the computer that answers for it.
- Web connection (HTTP)
- The way a browser asks a website for a page and the website answers.
- Container system (Kubernetes, K8s)
- Open-source software that runs an app’s parts across many servers and restarts them when they stop.
- Program-to-program connection (API, APIs)
- A way for one program to ask another for information or to do something, without a person clicking through a website.
- Plain-text data (JSON)
- A common way of writing data as named values, readable by people and programs alike.
- Response time (latency)
- How long something takes to answer, usually counted in milliseconds – thousandths of a second.
- Signed address-book answers (DNSSEC)
- A signature on a domain’s address-book answers, so a computer can tell they were not forged on the way.
- Secure web connection (HTTPS)
- The encrypted way a browser and a website talk, so nobody on the way can read or change what passes – the padlock in a browser.
- Content delivery network (CDN)
- Servers around the world that keep copies of a website so it loads quickly wherever you are.
- An address a program calls (endpoint)
- One web address of a service that a program sends its requests to.