agent-desktop
Pass
Audited by Gen Agent Trust Hub on Apr 7, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill provides instructions for legitimate desktop automation tasks on macOS using a sub-agent architecture.
- [COMMAND_EXECUTION]: The skill references various computer interaction tools (e.g., computer_launch, computer_click, computer_type) which are standard for UI automation. These tools are used according to the primary purpose of the skill.
- [EXTERNAL_DOWNLOADS]: The skill references a native binary agent-desktop. Based on the author context, this is a vendor-specific requirement for the skill's intended functionality.
Audit Metadata