ui-ux-pro-max
Pass
Audited by Gen Agent Trust Hub on Sep 30, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill executes entirely within the local environment using standard library Python scripts. It does not require or install third-party packages, nor does it perform any network operations to untrusted domains.
- [SAFE]: All search and recommendation logic is performed against static, locally bundled CSV files containing verified UI/UX best practices and technical specifications.
- [SAFE]: File system operations are limited to the creation of design system documentation in a dedicated local directory. The scripts include basic sanitization for file paths, such as replacing spaces with hyphens in project and page names.
- [INDIRECT_PROMPT_INJECTION]: The skill contains a surface for indirect prompt injection because it incorporates user-supplied strings, such as project names and search queries, into a 'MASTER.md' file that the agent is subsequently instructed to read as a source of truth for design rules. 1. Ingestion points: User input for project names and search queries in
design_system.py. 2. Boundary markers: Present via markdown headers and tables in generated files. 3. Capability inventory: The skill has local file-write capabilities for documentation purposes. 4. Sanitization: The skill performs basic string normalization for file names but interpolates content directly into the documentation bodies.
Audit Metadata