terminal-ui-engineer
Pass
Audited by Gen Agent Trust Hub on Mar 10, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The generated scripts utilize administrative privileges (sudo) to install system packages via apt and pacman.
- [EXTERNAL_DOWNLOADS]: The skill orchestrates the download and installation of the 'gum' utility from official package registries including Homebrew, Scoop, and Winget.
- [PROMPT_INJECTION]: The skill facilitates indirect prompt injection by incorporating user-defined functional goals into generated executable scripts without sanitization.
- Ingestion points: User-provided input for the 'Functional Goal' gathered via the question tool.
- Boundary markers: Absent from the generated script templates.
- Capability inventory: The generated scripts can perform file system operations (e.g., rm), execute build commands (e.g., npm), and run arbitrary shell logic.
- Sanitization: No input validation or sanitization is performed on the user-provided goal before code generation.
Audit Metadata