skills/cin12211/orca-q/ui-ux-pro-max/Gen Agent Trust Hub

ui-ux-pro-max

Warn

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: MEDIUMCOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION] (MEDIUM): The skill directs the agent to use run_shell_command to execute a local script (.shared/ui-ux-pro-max/scripts/search.py) with arguments generated from the user's prompt.
  • Evidence: The workflow (Step 2 and Step 3) repeatedly uses the pattern python3 .shared/ui-ux-pro-max/scripts/search.py "<keyword>" where <keyword> is extracted from user requirements.
  • Risk: If a user provides malicious input containing shell metacharacters (e.g., ;, |, `) and the agent extracts it verbatim as a keyword, it could lead to arbitrary command execution on the host system.
  • [Indirect Prompt Injection] (LOW):
  • Ingestion points: User-provided text describing UI/UX products, styles, and industries (File: SKILL.md).
  • Boundary markers: Absent. The skill does not provide delimiters or instructions for the agent to distinguish between user data and its own operational instructions.
  • Capability inventory: Shell execution (python3) and file creation/modification (implied in the "build/implement" actions).
  • Sanitization: Absent. No instructions are provided for the agent to validate or escape the extracted keywords before they are used in a shell context.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Feb 17, 2026, 05:31 PM