Security

How we protect your data and your customers' data. What we do, and what we never do.

Where data lives

Everything is encrypted on the way in and out, and at rest. Sign-in tokens for the apps you connect are kept in a separate encrypted vault. Test data and production data are kept apart.

Who can see it

Your workspace is yours. No other workspace can see it. Access by our staff is limited to what is needed to run the service, and every access to personal data is logged. All our provider accounts use strong passwords and two-factor sign-in.

What the assistant is allowed to touch

Only the tools for what you connected, and only at the trust level the customer has earned. A customer who has not proven who they are cannot reach any order, booking, or invoice. For billing we ask for a read-only key. The assistant never gets a key that can move money.

Least data

We ask each connected app for the smallest set of permissions the feature needs, and use them only for that feature. Files we fetch are deleted once read. Disconnecting a source deletes its content.

Documents customers send

Stored for your team. Not read by the assistant.

If something goes wrong

We have a written incident process: detect and assess at once, contain within hours, tell affected businesses within 72 hours of confirming a breach, and publish what we learned within two weeks.

Agreements

A data protection agreement that sets out our role as your processor is available on request. So is the incident response policy. Ask us.

Reporting a problem

Found a security issue? Write to us with "Security" in the subject. We read those first and we do not take action against good-faith reports.