design
Pass
Audited by Gen Agent Trust Hub on Oct 17, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill references a local Python script for information retrieval.
- Evidence: SKILL.md mentions 'searchable guidance via scripts/search.py' in the ui-ux-pro-max section for design systems and domains.
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied design briefs, creating a surface for potential instruction injection in untrusted data.
- Ingestion points: Design briefs (referenced in SKILL.md).
- Boundary markers: No explicit delimiters or instructions to ignore embedded content are specified.
- Capability inventory: Local script execution (scripts/search.py).
- Sanitization: No sanitization of brief content is documented.
Audit Metadata