winapp-ui-automation

Pass

Audited by Gen Agent Trust Hub on May 18, 2026

Risk Level: SAFEDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
  • Interaction with Application UI State: The skill allows the agent to read text values, inspect element trees, and capture screenshots of running Windows applications.
  • Evidence: Commands like winapp ui get-value, winapp ui inspect, and winapp ui screenshot provide deep visibility into the state of other applications.
  • Context: This is the primary purpose of the skill. While it involves accessing potentially sensitive information displayed on the screen, it is a standard requirement for UI automation tasks.
  • Indirect Prompt Injection Surface: The skill processes data from external applications (UI text and elements), which could theoretically contain instructions intended to influence the agent's behavior.
  • Evidence: The winapp ui get-value and winapp ui search commands ingest text directly from the UI of third-party apps into the agent's context.
  • Context: This is a common consideration for any skill that reads untrusted external data. The risk is minimized by the agent's internal guardrails, but users should be mindful when using the skill with applications displaying untrusted content (e.g., a web browser viewing an unknown site).
Audit Metadata
Risk Level
SAFE
Analyzed
May 18, 2026, 01:31 PM
Security Audit — agent-trust-hub — winapp-ui-automation