here are the answers to the questions you asked most under the Bot Screen launch:
- where it runs: on a Linux gateway host, like a server or a cloud VM. Hermes Desktop connects to it from your laptop, and each bot gets its own desktop there, not yours
- does the bot see what you type: no. while you hold control, its computer use and browser tools are refused, screen captures included. when you hand back, it re-captures the screen and carries on
- does it stop mid-task: when it reaches a login, 2FA, a CAPTCHA or a payment, it tells you in chat and ends its turn. take over, do the step, hand back, tell it to continue
- how heavy is it: plan on 1.1 to 1.5 GB of memory per open screen with a browser. a screen nobody uses stops after 30 minutes and comes back on the next use
- it is off by default: start it from the Screen pane in Hermes Desktop, or run `hermes computer-use screen start` on the host
You can now watch your agents work live as Hermes Desktop streams the screen of any bot's screen or session in real time.
Watch it drive a browser, type into a terminal, or open a window. Take over anytime to interact with the Bot Screen or type credentials and seamlessly hand back off when you're done.