1. Home
  2. Security & Data
Security & data

Your systems. Your data. Your rules.

AI workers touch real systems, so we keep them on a short leash: limited access, clear boundaries, a full activity log and a pause button that is always one click away.

WORKER-GUARDRAILS
01Connects with limited accessScoped
02Works inside set boundariesLimits
03Every action recordedLogged
04Risky actions wait for youYou approve
05Paused in one clickStoppable
5 guardrails · on every worker

WHAT WE PUT AROUND EVERY WORKER

How we connect

Only the access a worker needs.

Scoped access

Where a tool supports it, workers connect through limited, permission-based access instead of a shared password. They can only reach what the job needs.

Credentials stay protected

Access credentials are stored encrypted, and you can revoke a worker’s access at any time.

Worker boundaries

Clear rules for every worker.

Which systems

Each worker is limited to the tools you approve.

What data can move

You decide what it can read, write and send, and where.

Spending caps

Dollar limits are built into the worker, with no workarounds.

When it must ask

Defined actions always wait for a person.

Visibility

Everything a worker does is written down.

You can review what it did, when and why, and export the log for your own records. If you want to stop a worker, you can pause it immediately.

  • A reviewable, exportable activity log.
  • Clear flags on anything unusual.
  • Pause any worker, any time, in one click.
  • Your team can always take over by hand.

Your business data

Used to run your workers, not to train public models.

Your business content is used to run your own workers. We don’t use it to train public AI models, and we don’t sell it. Our Privacy Policy explains what we collect on this website and how we use it.

You stay in control

Three modes, set task by task.

New builds start in Draft and move on only when the pilot results support it and you agree.

Draft: the worker prepares, you decide.

The worker does the reading, checking and writing, then hands you a finished draft. Nothing goes out without your say-so. Every new build starts here.

BEST FOR: NEW BUILDS AND SENSITIVE TASKS
REFUND-REQUEST · DRAFT MODE
01Customer asks for a refund by emailAI worker
02Order found in your systemAI worker
03Refund policy checkedAI worker
04Refund decisionYou
05Refund issuedYou
06Customer told and ticket closedYou
6 steps · 3 prepared by the worker · 3 done by you

FAQ

Security questions we hear most.

Do you replace our login with a shared password?

No. Where a tool supports it, we connect through limited, permission-based access, and you can remove it at any time.

Can a worker spend money or send things without asking?

Only inside limits you set. Anything beyond them waits for a person.

Can I see what a worker did?

Yes. Every action is logged, and the log can be reviewed and exported.

Is our data used to train public AI models?

No. Your business content is used to run your own workers.

Who do I ask security questions?

Email [email protected] or call 407.410.4420, and a person will answer. You can ask before you book an assessment.

Ready when you are

Questions before you book?

Ask us anything about access, data or controls, before you commit to anything.