ui-ux-pro-max
Warn
Audited by Gen Agent Trust Hub on Nov 7, 2026
Risk Level: MEDIUMPRIVILEGE_ESCALATIONCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [PRIVILEGE_ESCALATION]: The skill provides instructions for the agent to use
sudoto install the Python runtime on Linux-based systems, which requires escalating privileges to the system administrator level. - [COMMAND_EXECUTION]: The skill operates by executing a local Python script (
scripts/search.py) via the shell to generate design recommendations and persist data to the file system. - [INDIRECT_PROMPT_INJECTION]: The skill exhibits an injection surface where user-supplied product descriptions and keywords are placed directly into shell command arguments (
python3 ... "<query>") without explicit sanitization or validation. It also retrieves context from local files (design-system/MASTER.md) that could be manipulated to influence agent behavior. - Ingestion points: User input for query parameters and local design system markdown files.
- Boundary markers: No clear delimiters are specified to isolate untrusted input within the command templates.
- Capability inventory: The skill has the ability to execute shell commands and perform file write operations for design persistence.
- Sanitization: There are no instructions for escaping or validating user input before it is used in subprocess execution.
- [EXTERNAL_DOWNLOADS]: The skill facilitates the installation of the Python 3 runtime from official and well-known operating system package registries (such as Homebrew, APT, and Winget).
Audit Metadata