uninstall

Pass

Audited by Gen Agent Trust Hub on Oct 6, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes a local build script (.\build.ps1) and a locally generated binary (OpenClaw.Tray.WinUI.exe). These actions are used to manage the lifecycle and cleanup of the application's developer environment, aligned with the skill's primary intent.
  • [INDIRECT_PROMPT_INJECTION]: The skill captures and displays the output of the uninstallation process for user verification.
  • Ingestion points: Reads output and error logs (out.log, out2.log) generated by the uninstallation executable in the local filesystem.
  • Boundary markers: None implemented in the script; the skill instructs the user to manually review the logs for specific verification strings.
  • Capability inventory: The skill utilizes capabilities to terminate processes (Stop-Process), unregister WSL distributions (wsl), and delete application data in %APPDATA%.
  • Sanitization: No automated sanitization or filtering is applied to the log content before it is presented to the agent context.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 6, 2026, 11:37 AM