See the actions.
Shell commands, file changes, tool calls, network requests, and browser interactions — in one activity trail.
Security workspace · Early access
The monitoring and safety layer for everything your agents do — across desktop and browser.
Shell commands, file changes, tool calls, network requests, and browser interactions — in one activity trail.
Destructive commands, unexpected access, and suspicious instructions get flagged. Skills are checked for security risks when you pull them.
Token consumption by agent and session, wherever the harness exposes it.
We’re onboarding a small group first. Join to help shape v1 and get early access.