windows-qa-engineer

Pass

Audited by Gen Agent Trust Hub on Mar 11, 2026

Risk Level: SAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The setup process involves cloning the UFO framework directly from Microsoft's official GitHub repository (github.com/microsoft/UFO). As Microsoft is a trusted organization, this is considered a safe and necessary dependency for the skill's primary functionality.- [PROMPT_INJECTION]: The skill exposes an attack surface for indirect prompt injection because it reads text content from the User Interface of the application under test.
  • Ingestion points: UI text is retrieved from the system via the 'texts' and 'get_app_window_controls_info' tools defined in 'scripts/ufo_windows_qa_mcp_server.py'.
  • Boundary markers: No specific delimiters or safety instructions are implemented to prevent the agent from accidentally executing instructions found within the UI text of a tested application.
  • Capability inventory: The skill allows the agent to interact with the host system's UI through 'click_input', 'set_edit_text', and 'keyboard_input' via the 'AppUIExecutor' server.
  • Sanitization: The skill does not perform sanitization or validation of the text retrieved from the UI before it is processed by the agent.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 11, 2026, 12:31 AM