Security
Security that watches all the time and acts on its own
Sign-ins, holes, leaked keys, lookalike domains and spoofed email – watched and acted on.
- Your device 10.0.0.5 and 203.0.113.9 on the internet exchanged traffic that matched "ET MALWARE Observed Cobalt Strike CnC Domain" 3 times between 2026-10-05 11:50 and 2026-10-05 11:52 UTC. It is malware contacting its controller. Nothing blocked it - your sensor watches and reports.
- In attack terms, a device is talking to a server an attacker controls (MITRE ATT&CK T1071, Application Layer Protocol). Risk 90 of 100 (critical): its authors mark it critical (+15); the signature is high-confidence (+5); a device inside your network is the one talking out, so it may already be compromised (+15).
- Block 203.0.113.9 on your firewall - Xovris shows the exact change as a dry run first, and it ends by itself after the time you choose unless you keep it.
- Switch this signature to block on your sensor (IPS mode) - also a dry run first, undone with one press.
The first of the 3 alerts, as the sensor sent it
{"timestamp":"2026-10-05T11:50:12.123456+0000","flow_id":1234567890,"in_iface":"eth0","event_type":"alert","src_ip":"10.0.0.5","src_port":51234,"dest_ip":"203.0.113.9","dest_port":443,"proto":"TCP","app_proto":"tls","hos…XovrisOn its own
- Xovris keeps watching this incident, tells the people on call, and prepares the block below as a dry run so one press makes it safe.
- the sensor rates the signature most severe (55)
Sign-ins from new places, holes in devices and software, leaked keys, lookalike domains, spoofed email, intrusion alerts and the sites your people visit, for example. Xovris watches each one, acts on its own where your policy allows and shows the proof.
Intrusion detection
Alerts from Suricata, Zeek, Snort 3, Wazuh, FortiGate and Palo Alto read into one record with a shared flow ID, the rule’s own metadata and the MITRE ATT&CK step it belongs to, then grouped into incidents so one scan is one thing to read, not a thousand.
- Suricata EVE, Zeek, Snort 3, Wazuh, FortiGate IPS and PAN-OS threat logs read into one record with a shared Community ID
- Each signature’s own metadata and its MITRE ATT&CK step, with the CVEs it names
- Alerts from one source in one window grouped into one incident, ranked by risk, with what to do
XovrisOn its own
- Reads every alert into one record
- Names the attacker’s step
- Groups a scan into one incident
The web filter for every office and home
Known-bad, phishing and lookalike sites blocked at the name, from open lists, with categories you choose for the office and age-right profiles for a family. A site nobody has listed yet is judged as it is asked for.
- Phishing, malware and lookalike names blocked from licence-checked open lists (UT1, URLhaus, Phishing.Database)
- A name never seen before judged as it is asked for, from how it is made and what it imitates
- Categories aligned to Cloudflare Gateway for the office, age-right profiles and safe search for a family
XovrisOn its own
- Blocks known-bad names
- Judges a name nobody listed
- Applies your categories
Holes in devices and software
Every asset’s software matched to known vulnerabilities and ranked by what matters: exploited now, likely to be exploited soon, then severity and exposure, with the one upgrade per package that fixes the most.
- Every device you connect, checked for known security holes, worst first
- 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
- A broken build fixed and proven before CI runs again, and every change checked for attack paths
- The AI review runs on every paid plan
- Your code's bill of materials in the command centre
- See what people did beside what Xovris did
- Know which vulnerable packages are actually served
- See who may do what at a glance
- Password health across the company
- Travel mode for laptops and phones
- Insider risk, lawfully and openly
- Secrets and personal data masked, field by field
- Your data masked before any AI reads it
- Alerts when a key is used in a way it never was
- What you delete stays deleted
- Containment lifts itself when it is safe
- Containment checked where the threat was
- A containment is judged by what it stopped
- Quarantine lifts only once it is clean
- Containment rehearsed before it is used
- Containment never takes your other services down
- Sign-in problems explained from the source
- Vulnerabilities read from the authority's own records
- An approval runs only the exact change you approved
- One emergency stop for everything
- Every access decision explained in plain words
- Payment requests by voice or video held for checking
- Session replays masked by default
- Feedback reaches its own session and no other
- Masking proven everywhere a session is seen
- Xovris acts only on messages from people still in your team
- Every action from chat is re-checked at the moment it runs
- Security headers set on any web server
- Identity steps done for you
- Everything Xovris is built from is pinned
- Your data walled off from every other workspace
- Nothing you send is ever used to train AI
- Every action Xovris takes is on a tamper-evident record
- Decoy accounts and documents that catch intruders
- New tools run sealed off the first time
- Mail records checked beside every DNS repair
- DMARC read to the newest standard
- A DMARC policy of 'none' moved on for you
- Inbound events checked against replays
- Chat replies act as the person who sent them
- Xovris acts only for people in your team who may act
- Every inbound callback authenticated
- Expiring keys rotated before they lapse
- Single points of failure in people, found
- Any security product connects, not only the five listed
- Every repair checked against the access you granted
- Apps with unusual access to your data, found
- Certificate fixes rehearsed with a real certificate authority
- Every change passes one safety gate
- Nothing is lost, and backups are proven by restoring them
- The check and the fix say exactly the same thing
- Lost your authenticator? Xovris hands you the way back
- A locked account is never made worse
- Tool searches carry nothing of yours
- Every tool's licence is read first
- Abandoned and look-alike packages are refused
- Tools with known vulnerabilities are refused
- Every tool is checked byte for byte
- Tools are removed when the job is done
- DNS records left pointing at nothing are removed
- Weak or missing DKIM keys found at your mail provider
- A certificate that does not match is reissued
Sign-ins and accounts
An account at risk is locked down, passkeys can be required and a leaver loses every account at once.
- Password-only sign-ins found, and an affected account locked down in one step
- Require a passkey for everyone in your workspace
- Know whether a leaked API key is switched off for you
- Your keys never pass through the AI
- Your keys stay sealed, every step stays inside what you allowed, and nothing Xovris reads can tell it to act
Scams, lookalikes and breaches
Is this message, link or site a scam? answered with the evidence. Your own addresses, domains and brand watched for lookalike domains and breaches through lawful sources, with what to do now.
PayPal Support <alert@secure-paypa1-help.com>
Urgent: your account will be suspended
Please verify your account within 24 hours or it will be closed.
www.paypal.comhttps://paypa1.com.login-secure.xyz/verify
- Do not click its links, open its attachments or reply.
- If it claims to be a company you use, contact them through their own app or the address you already know.
XovrisOn its own
- A scam
- it asks for a password, a PIN or a code - a real company never asks for these
- it says it is from PayPal Support but was sent from secure-paypa1-help.com
- the address imitates "paypal" as "paypa1", a look-alike letter swapped in, and really belongs to login-secure.xyz
- the link says www.paypal.com but goes to login-secure.xyz
- “Is this a scam?” answered with the reasons – and what to do next if you already clicked, typed a password or paid
- Your email addresses, domain and brand watched for breaches and lookalikes, and the response done for you
- A lookalike site's abuse report is filed by Xovris the moment it can be, never left to you
- A lookalike page or a forged link never becomes a fact about your incident
- Four more free checks: blocklists, security headers, Cyber Essentials readiness and leaked passwords
Compliance from live evidence
Cyber Essentials v3.3 judged requirement by requirement from what Xovris already sees, each one met, not met or unknown with its evidence and its fix.
- UA-1Every sign-in to a cloud service always uses two-step sign-in (MFA)finance (shared mailbox) on Microsoft 365
- FW-1Every firewall's default admin password is changed, or remote admin is off
- FW-2No firewall admin page is open to the internet unless MFA or an allow list protects it
- FW-3Inbound connections are blocked by default and every inbound rule is documented
- SC-1Every device needs a password, PIN or biometric to unlock, at least 6 characters
- SC-2Unlocking is protected against guessing: no more than 10 tries
- SC-3Auto-run is off and unneeded accounts are removed
- SU-1All software is licensed and in its maker's support - anything past its support is removed
XovrisOn its own
- 14 of 15 requirements met, 1 not met - one would fail the assessment automatically.
- Turn on two-step sign-in for these accounts - Xovris walks each person through it in minutes; a cloud service that offers none has to be replaced.
Let Xovris watch for you
Start free and Xovris begins with your own domain and email the same minute.
Words on this page, in plain English
Each technical word this page uses, said simply.
- 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.
- Email rules record (DMARC)
- An address-book record telling mailboxes what to do with email that fails the domain’s sender checks – deliver it, mark it as spam or refuse it – and where to send reports.
- Email signature (DKIM)
- A signature added to each email that proves it came from the domain and was not changed on the way.
- 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.
- Customer space (tenant, tenancy)
- One customer’s own separate space in a shared service, kept apart from everyone else’s.