ui-ux-pro-max
Pass
Audited by Gen Agent Trust Hub on Sep 28, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to extract keywords from user requests and interpolate them directly into a shell command (
python3 .shared/ui-ux-pro-max/scripts/search.py "<keyword>"). - Ingestion points: User-provided requirements such as product type, style keywords, and industry (Step 1 in
SKILL.md). - Boundary markers: Absent. There are no instructions or delimiters provided to prevent the agent from including shell metacharacters in the interpolated command.
- Capability inventory: Use of
run_shell_commandto execute a local Python script (Step 2 inSKILL.md). - Sanitization: Absent. The instructions do not mention escaping or validating the extracted keywords before execution.
- [COMMAND_EXECUTION]: The skill's primary functionality relies on executing a local script via the shell.
- Evidence: Instructions in
SKILL.mddirect the agent to runpython3 .shared/ui-ux-pro-max/scripts/search.pymultiple times during its workflow.
Audit Metadata