The platform

Eclipse

Our own model, trained to attack.

It works on its own, day and night.

Cantic console · demo data

An illustration with demo data: Eclipse tries each way in that is in scope, proves the one that works with a step-by-step replay, and the fix is written, applied by your IT provider and retested.

The difference

Scanners report what might be wrong. Eclipse reports only what it can demonstrate.

Our own AI attacker

It attacks the way a criminal would.

It maps what you expose

Everything your company has reachable from the internet.

It tries every way in

From the website and email to the systems that run your plant and cold store.

It proves every finding

When it gets in, it proves it and measures the risk.

Proof, not noise

Every finding comes with proof.

A step-by-step replay

Every action is hashed, timestamped and kept in a ledger you can inspect, and you can follow a test live.

Dispute a finding

Think a finding is wrong? A person on our team will re-check it. Dispute a finding

A named lead on every critical finding

Each critical finding goes to a named lead on our team.

Cantic console · demo data

Finding · criticalProven

Contractor portal open to the internet, default password accepted

Step-by-step replay

09:15:41Portal found at 198.51.100.52:443#7f3a91

09:15:58Default password accepted#c2e04b

09:16:12Supervisor page reached#91d7a0

09:16:30Replay saved, every step hashed#0b4e6c

On the record

Named leadAssigned

DisputeOpen to you

Under your control

Inside the rules you sign.

Signed rules

Nothing runs until someone who can authorise testing, usually the owner or managing director, signs the authorisation and rules of engagement.

Proof the domain is yours

For recurring testing, you prove it with a DNS record or a file upload.

Rate-limited and time-boxed

It stays inside the scope, hours and limits you set, and stops once it can prove control.

Stop it within 60 seconds

You can stop a running test at any time.

Our own model, on our own servers

What a test gathers goes to no outside AI company.

Destroyed when the work is done

We destroy what a test gathered on our side, through a multi-step deletion.

Encrypted, under NDA

Findings are encrypted and handled under NDA. No client is ever named.

First test reports for 14 days

Kept for 14 days, unless you subscribe.

In a food company

What Eclipse looks at in a food company.

From the outside: Eclipse

Website, email, portals, VPNs, remote desktop, contractor logins and management pages.

The rest: Fortify

We review production lines to refrigeration, and plan for the day systems go down.

What we test, system by system

Plans

Keep Eclipse attacking for you.

Compare plans

Every week

Core

  • An attack run every week
  • Alerts on new flaws that affect you
  • Supplier advisories

All year

Fortify

  • At least one full test a month
  • The fix within 24 hours
  • A continuity plan

See what a criminal would see.

Your first test runs under rules you sign, with the findings in plain words, in one day.