Security / If we were broken into
If TenantWard were broken into
In October and November 2026 two tools that IT providers trusted with access to their clients’ Microsoft 365 were broken into, and the attackers used that access in hundreds of companies. Any tool with admin access carries the same risk, TenantWard included. This page says what someone who broke into TenantWard could reach, how we make that hard, how you cut TenantWard off in a minute, and what we would do.
What they could reach
TenantWard’s rules (a plan someone approved, a backup first, a fixed list of changes) bind TenantWard’s own code. Someone holding TenantWard’s keys directly wouldn’t go through those rules, so this is what the access itself allows, in every company that granted it:
- A read-only connection: read the people list, groups, licenses and admin roles, sign-in rules and methods, sign-in and audit logs, mailbox settings such as forwarding, outside apps, devices and sharing settings. Not the contents of mail or files.
- Change access: everything above, plus the changes listed on what TenantWard can do with your access: sign-in rules, adding and changing people (including resetting a password or MFA), groups, mailbox settings such as forwarding, and sharing limits. Admin roles and outside apps only where you added those permissions.
- Response access, where a Global Administrator approved it: block accounts and end sessions, and read and move mail in any mailbox. TenantWard uses the mail permission only to move a reported phishing email, but someone holding it wouldn’t be limited to that. If you don’t use automatic response or phishing removal, don’t approve it.
- An IT provider’s partner sign-in: act in that provider’s customers, within the admin roles each customer granted the provider.
- Google Workspace: what the scopes on the security page allow, which are read-only apart from the group, calendar and mailbox-forwarding changes TenantWard puts back.
- TenantWard’s own records: check results, change history, settings backups and the people list from the last check. TenantWard holds no passwords, and no mail or files.
How we make that hard
- TenantWard’s main Microsoft apps prove who they are with a certificate, not a password. Its private key is kept encrypted in the database’s vault, and one server function uses it to sign a ten-minute sign-in, only for a directory TenantWard is connected to. The key itself never leaves that function.
- An IT provider’s partner sign-in is kept encrypted in the same vault and used only for the customers the provider picked.
- Start read-only, and add change or response access only when you use them. Each optional permission can be left out or taken away on its own.
- Every call TenantWard makes to Microsoft 365 is logged in TenantWard, and Microsoft logs each of TenantWard’s sign-ins in your own directory too (Entra admin center, Enterprise applications, TenantWard, Sign-in logs), where we can’t change them.
- The staff console sits behind a separate sign-in with two-step verification, and every staff action in it is recorded.
- If our keys were ever taken, removing them from TenantWard’s app registrations stops every TenantWard sign-in, in every company, at once. That is the first step of our response plan below.
Cut TenantWard off in a minute
You don’t need us, or to be signed in to TenantWard, to do this:
- Microsoft 365: in the Entra admin center, open Enterprise applications and search for TenantWard. For each app whose name starts with TenantWard, open Properties, set “Enabled for users to sign in?” to No, and save. New sign-ins stop straight away, and a sign-in already issued stops working within about an hour. Set it back to Yes to carry on, without approving anything again.
- An IT provider’s partner access: end the delegated admin relationship in the Microsoft 365 admin center (Settings, Partner relationships), or the provider ends it in Partner Center.
- Google Workspace: in the Admin console, open Security, API controls, Domain-wide delegation, and delete TenantWard’s entry.
What we would do, and when you would hear
- Remove our own keys first, so TenantWard’s access stops everywhere while we find out what happened.
- Email the owners of every account within 24 hours of confirming it: what was reachable, in which of your companies, what we know was done, and what to check. Then an update every day until it’s closed.
- Post the same on the status page, and the full report when it’s over.
- Turn our access back on only after new keys are in place, and tell you before we do.
Found a weakness in TenantWard? Email hello@tenantward.com; the rules are under reporting a security issue.