How your data is kept, what we ask of you, and where to report a problem.
How it is built
What happens today, plainly.
Each one describes how Sentinel runs now, not a plan.
Access is granted one person at a time
For a domain whose ownership you have confirmed, a named person at your business can sign in to see their own domains. The Fourths grants that access one person at a time. If the contact is given the right, they can also answer 'whose sender is this?'
On the report, your IT makes the change
On the Monthly security report, we give your IT or provider the exact record and the steps. They publish it. For a record handed over in Sentinel, it re-checks every day until it shows.
The free check sends nothing on its own
The free domain check scans the domain behind the work email you type, shows your score and two findings on screen, and sends no email at that point. We keep the address, domain and score for 12 months from your last activity.
Each client’s data is kept apart
Every client table is filtered by row-level security in the database itself, not only in the application. A required check tests that on every change.
Sign-in through Microsoft
Staff and client contacts sign in with Microsoft Entra, and repeated sign-in attempts are rate limited.
Credentials kept in a vault
The app's credentials live in a managed key vault and reach it by reference, and every change is scanned for secrets before it can merge.
Every change is checked first
No change reaches the product until its tests, the isolation proof, a secret scan and a dependency scan have passed. Releases are made by hand.
Reports are not kept for ever
DMARC reports are deleted twelve months after they were generated.
What it needs from you
What we ask for, and what we do not.
Monitoring asks for no password and no access to your DNS or your mail. You publish the records yourself.
The services that receive data while Sentinel runs are named in our privacy policy, and a release check stops an optional service being switched on before it is disclosed.