One approval by a second person for the computer and its sites

The idea

One approval by a second person (an owner, or an admin who did not issue the computer) binds the computer's install manifest and allowed hosts; onboarding runs as a resumable job with a doctor, snapshot and restore, and each added site needs another such approval.

The problem
Installing tools and opening network access silently is unsafe.
Why it matters
A second person sees exactly what is installed and reached, and the issuing admin cannot approve it.
Where it stands
Built, off in production as of Shown on a real guest computer on a local backend in owned runs on November 14 and 15, 2026; production keeps it switched off.

Recorded evidence

  • Local run on the real backend · 19/19 gates through the product path
  • Local run on the real backend · An added site approved by a second person; the issuer could not approve it; the proxy changed and was read back, 20/20

Recorded walkthrough: The AI worker's own computer. It shows part of this idea: a second person, here the owner, approves one computer and its hosts, then the onboarding job runs its doctor, checkpoint and restore. The approval of an added site is not in these runs.

  • Worker's private persistent computer

    An admin issues a worker its own Linux computer that persists like a laptop until revoked, offboarded or expired, with full control inside its guest, an approved egress allowlist and isolation at the guest boundary (today a Docker container for a single tenant).

    Recorded walkthrough

  • Worker's own browser

    The worker drives a real browser on its own computer through governed tools and can check its own app on that computer; the platform treats every page as outside content, and admins see the worker's screenshots.

Turn the numbers into shipped work.

Crewshal runs these choices for you: a persistent AI worker with memory and rules, on your Claude and Codex subscriptions.