skills/jetbrains/skills/screenshot/Gen Agent Trust Hub

screenshot

Pass

Audited by Gen Agent Trust Hub on May 6, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill relies on several system-level utilities (screencapture, scrot, gnome-screenshot, import, osascript, swift, and defaults) to perform its core tasks. These commands are executed via standard subprocess calls across all supported platforms.
  • [COMMAND_EXECUTION]: In scripts/take_screenshot.py, the activate_app function uses string interpolation to build an AppleScript command for osascript. While it includes a sanitization step (escaping double quotes), this is an entry point where application names could influence script execution.
  • [COMMAND_EXECUTION]: The Windows helper scripts/take_screenshot.ps1 uses the Add-Type cmdlet to compile C# code at runtime. This code interfaces with the Windows API (user32.dll) to determine window bounds and focus, which is a common and legitimate pattern for advanced PowerShell automation.
  • [COMMAND_EXECUTION]: The PowerShell script instructions recommend the use of the -ExecutionPolicy Bypass flag to ensure the script runs regardless of the local environment's restrictive execution policies.
  • [SAFE]: All external calls and tool usage are strictly limited to the stated purpose of capturing and saving screenshots. The skill does not perform any network operations, and its file access is limited to standard user directories (Desktop, Pictures, Temp) as per user or agent request.
Audit Metadata
Risk Level
SAFE
Analyzed
May 6, 2026, 10:14 AM
Security Audit — agent-trust-hub — screenshot