wrenhouse

Security and trust

Your business stays your business.

Your hub holds your customers, your messages and how you work. Here is how we look after it, in plain words.

You control what goes out

Every automation starts by asking a person first. Nothing is sent, booked, charged or changed in another system on its own until your admin turns on automatic mode for that one automation.

Every hub is kept apart

Each customer's hub, people and history are separate. Our tests check on every change that one customer can never see another's hub.

Support access is logged

If our team opens your hub to help, we have to give a reason first. The reason, the time and who it was are written to your hub's log.

Your data doesn't train AI

Agents run on Claude by Anthropic through its business service. Under Anthropic's commercial terms, what your hub sends is not used to train its models.

Hosted in the United States

Wrenhouse runs on Render, a hosting provider with its own SOC 2 Type II report. Traffic is encrypted with HTTPS, and databases are encrypted at rest.

Sign-in done properly

Passwords are stored only as strong one-way hashes. Invite codes work once and expire. Sessions use secure cookies, and requests from other websites are refused.

Learning is read-only

Your hub learns only from the sources your super admin turns on. It reads, it never changes them, sensitive details are masked, and every read is logged. You can correct or delete anything it learned.

Every change can be undone

Every version of your hub's design is kept. If something changes that you didn't want, it rolls back in a click.

Your data stays yours

We use it only to run and support your hub, and we never sell it. When you leave, ask for a copy and we delete the rest.

Tell us about a concern

If you think you've found a security problem, write to support@wrenhouse.io with "Security" in the subject. A person reads every one.

Questions from your IT team?

The full Security and Trust page lists our hosting, every provider we use and each permission your tools grant. We're also glad to walk your team through it and fill in your security questionnaire.