Skip to content
xovris.ai · 5 of 5 fine · 2 Oct, 01:49 UTC
Join early access
Menu

Trust centre

Everything Xovris can fix, and every control you keep

What Xovris can fix, how deep it goes and how every repair is proven – with every control you keep.

What Xovris did on customers' systems

Live

Counted from the audit record of every change, across all workspaces together, so no customer can be picked out.

The live record of what Xovris’s AI does opens at launch, with every figure counted from real work.

Xovris right now

Xovris’s live status opens at launch, measured by its own probes from the first day.

How deep Xovris goes

From the page your visitors see down to the networks and devices behind it. Each layer shows what Xovris fixes there and whether it is ready today.

  1. Websites and pages

    Available

    Pages that fail, slow down or go offline – found from outside, fixed and proven for every visitor.

    See it working
  2. Domains and DNS

    Beta

    Changed or tampered records put back to their last verified version, with expiry and renewal watched.

    See it working
  3. Email and sender trust

    Available

    Who sends email as you, and the exact records that stop anyone else from doing it.

    See it working
  4. Apps and code

    Beta

    Errors traced to the release that caused them, and the fix delivered as a change you review.

    See it working
  5. Security

    Beta

    Certificates, DNSSEC and tampering caught early and put right.

    See it working
  6. Cloud accounts

    Planned

    Who changed what in your cloud, and a risky setting put right.

  7. Networks

    Planned

    Office and home networks watched, and a fault traced to the device behind it.

  8. Devices

    Planned

    The laptops, phones and machines a business or a family relies on, kept healthy.

Every repair is tested before and after – and guaranteed

A fix counts only when it is proven. Until then you pay nothing for it.

  1. Reproduced first

    Your own failing outcome is reproduced before anything changes.

  2. Fixed within what you allow

    Only the access you granted, with every step recorded.

  3. Proven after

    Re-run after the fix with the checks around it, and confirmed by a verifier separate from the fixer.

  4. Paid when it passes

    You pay nothing for a fix until its after-test passes.

If an after-test fails, the change is put back exactly as it was and the next approach starts. The issue stays open and visibly progressing until a fix is proven.

The only limits: law, safety and consent

Xovris goes as deep as your systems go. These three are the only things that stop it, and each is named when it applies.

What each connection lets Xovris fix

Each provider in turn: the fixes it unlocks first, then the exact access its own consent screen asks for and how to take it back – read access shown apart from change.

Connections not listed here – the SDKs, OpenTelemetry and the coding-agent server – send data to Xovris and get no access to your systems at all.

Certifications

Xovris holds no certification yet. Each is listed with its true status, and gains its date and its evidence the day it is earned – never before.

Your data, and the AI

Where your data lives

Your data lives in the EU (Netherlands and Germany) site, chosen when you sign up.

  • Your account, checks, incidents and changes: the Netherlands
  • Error events and check results: the Netherlands
  • The history of each repair while it runs: Germany

Your data stays in the site you choose. Moving it later means exporting it and importing it into another site.

Sub-processors

None yet. Xovris processes no customer data until it opens, and every sub-processor that will handle your data, and where, is listed here before it does.

AI governance

  • Your data never trains anyone else’s AI. Xovris learns from fixes only after names, addresses, secrets and content are removed, and only as your settings allow. Xovris only uses providers whose terms forbid training on customer content.
  • A person approves every change unless you set a standing policy for that kind of change.
  • Anything written by Xovris is labelled as written by Xovris.
  • Nothing from one workspace is ever visible to another – the database itself enforces it.
See what each level of autonomy asks for

Report a vulnerability

Write to [email protected] with “Security” in the subject. Please do not test against other customers' data or degrade the service.

If it is being exploited now, put “Exploited” in the subject as well. Those reports are handled first, and an actively exploited vulnerability in Xovris is reported to ENISA within 24 hours, with an update within 72 hours and a final report, as the Cyber Resilience Act requires.

  • What you found and where
  • How to reproduce it, step by step
  • What an attacker could do with it

[email protected] · security.txt

Accessibility

Xovris is built to WCAG 2.2 level AA.

Not yet independently audited; the result will be published here with its date.