ui-ux-pro-max
Pass
Audited by Gen Agent Trust Hub on Jun 17, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to use
run_shell_commandto execute a script located atshared/ui-ux-pro-max/scripts/search.py. This script takes keywords directly from user input to search for design patterns.- [PROMPT_INJECTION]: The skill presents an indirect prompt injection surface because it takes keywords extracted from the user's request and interpolates them directly into a shell command argument without explicit instructions for sanitization or escaping. - Ingestion points: User-provided project requirements and style keywords extracted in Step 1 of the workflow.
- Boundary markers: There are no boundary markers or instructions to treat the interpolated keywords as untrusted data.
- Capability inventory: The skill uses
run_shell_commandto execute Python scripts, which provides a direct vector for command injection. - Sanitization: The skill lacks any instructions to sanitize, validate, or escape shell metacharacters before executing the search command.
Audit Metadata