ui-ux-pro-max
Warn
Audited by Gen Agent Trust Hub on Jun 20, 2026
Risk Level: MEDIUMCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The persistence logic in
scripts/design_system.pyis vulnerable to path traversal. Thepersist_design_systemfunction constructs file paths using user-supplied parameters (--project-nameand--page) without sanitizing for directory traversal sequences (e.g.,../) or absolute paths. This allows the skill to write markdown files to arbitrary locations on the host filesystem when the persistence feature is used. - [COMMAND_EXECUTION]: The skill requires the agent to execute a local Python CLI tool (
scripts/search.py) with arguments derived from user input. While the script is local, it accepts complex strings as arguments, presenting a surface for attack if the agent fails to properly escape shell metacharacters during execution. - [PROMPT_INJECTION]: The skill has an indirect prompt injection surface (Category 8).
- Ingestion points: User-provided product types, industry keywords, and style requirements enter the agent's context and are interpolated into design system generation logic.
- Boundary markers: There are no delimiters or explicit instructions to distinguish untrusted user requirements from the skill's internal reasoning data.
- Capability inventory: The skill can write files and generate structural guidelines that significantly influence the agent's future code generation tasks.
- Sanitization: No validation or filtering is performed on user requirements before they are incorporated into the generated design system instructions.
Audit Metadata