FAQ

ClawPaw operates at the operating system level. That naturally raises practical questions about privacy, control, and behavior. The following clarifications address common concerns.

Does ClawPaw send my files to the cloud?

ClawPaw runs locally. Your files remain on your device unless you explicitly choose to use an external LLM API. System context is not mirrored to a remote virtual desktop or hidden infrastructure layer.

ClawPaw does not operate a background cloud replica of your environment.

Can ClawPaw access my entire computer?

No. ClawPaw can only access:

• Folders you explicitly approve
• System services you grant permission to
• Context relevant to the active task

It cannot bypass macOS privacy controls or escalate privileges.

Does it monitor my activity continuously?

No. OS-level awareness is invocation-based. Context is read when you trigger ClawPaw, not continuously harvested in the background.

ClawPaw does not record browsing history, track application usage patterns, or collect behavioral analytics.

Can ClawPaw act without my approval?

Autonomy is adjustable per task. In Explore and Ask modes, no system changes occur without visibility. Execute Mode allows direct action, but only within granted boundaries.

Intelligence does not override authority.

How is this different from developer agent frameworks?

Framework-based agents typically require API configuration, scripting, and infrastructure setup. ClawPaw is designed for direct interaction within your operating system.

You do not configure the system first. You describe intent first.