ui-ux-pro-max
Fail
Audited by Gen Agent Trust Hub on Feb 26, 2026
Risk Level: HIGHCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill relies on executing a local Python script at
skills/ui-ux-pro-max/scripts/search.pyto perform reasoning and search operations. - [COMMAND_EXECUTION]: Instructions explicitly direct the agent/user to run
sudo apt update && sudo apt install python3on Linux systems, which involves elevated administrative privileges. - [PROMPT_INJECTION]: The skill implements a hierarchical retrieval pattern that directs the agent to read and prioritize rules from dynamically generated local files like
design-system/MASTER.mdanddesign-system/pages/*.md. This creates an indirect prompt injection surface where external inputs processed by the search script can influence future agent behavior. - Ingestion points: User-provided keywords and product descriptions passed as CLI arguments to the search script.
- Boundary markers: Absent; the agent is simply told to "read design-system/MASTER.md" and "prioritize its rules."
- Capability inventory: Subprocess execution of Python scripts and filesystem write/read operations.
- Sanitization: Not specified for the content written to the design system files.
Recommendations
- AI detected serious security threats
Audit Metadata