ui-ux-pro-max
Pass
Audited by Gen Agent Trust Hub on Jul 28, 2026
Risk Level: SAFE
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes a local Python script (scripts/search.py) to lookup design heuristics. The agent is instructed to execute this script via the command line to generate responses based on user requirements. This is standard behavior for skills using helper utilities.
- [INDIRECT_PROMPT_INJECTION]:
- Ingestion points: User-specified product types and keywords are passed as command-line arguments to the internal search script via the instructions in SKILL.md.
- Boundary markers: The instructions recommend wrapping user-provided tokens in double quotes when forming the execution command.
- Capability inventory: The skill can execute its bundled script and write markdown-formatted design system files to the local workspace using standard file I/O.
- Sanitization: The script uses the standard argparse library for argument parsing; while it does not explicitly filter for path traversal characters in the optional file persistence argument, the write operations are restricted to the local directory and enforced with a .md extension.
Audit Metadata