dev-test-linux

Pass

Audited by Gen Agent Trust Hub on Jun 22, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: Provides extensive bash commands and scripts for automating UI interactions using xdotool, ydotool, wtype, and dbus-send.
  • [EXTERNAL_DOWNLOADS]: Recommends installing system utilities (ydotool, wtype, grim, slurp, xdotool, xclip, scrot) from official Linux distribution repositories via apt, pacman, and nix-env.
  • [PROMPT_INJECTION]: Implements logic to search for windows by title and traverse UI accessibility trees, creating a surface for indirect prompt injection.
  • Ingestion points: Reads window names (xdotool search --name) and accessibility tree metadata (pyatspi) from the operating system environment.
  • Boundary markers: Absent; the skill does not define specific delimiters or instructions to ignore embedded content in UI element data.
  • Capability inventory: Includes comprehensive shell execution for UI automation, input simulation, and D-Bus service control.
  • Sanitization: Absent; UI metadata is used directly in search operations without escaping or validation.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 22, 2026, 06:30 AM
Security Audit — agent-trust-hub — dev-test-linux