no-ui-slop
Pass
Audited by Gen Agent Trust Hub on Oct 12, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted or external data which could potentially contain malicious instructions.
- Ingestion points: Reads local UI code, design systems (DESIGN.md), and external UI references/materials fetched via the
find_ui_referencesandfind_ui_materialstools. - Boundary markers: Instructions advise the agent to learn design patterns from references but to fit them specifically to the target product, which helps separate the influence of external data.
- Capability inventory: The skill is restricted to UI design tasks and does not have access to subprocess execution, file-writing outside UI context, or general network operations.
- Sanitization: No explicit sanitization or filtering of external design data is defined; the skill relies on the underlying model's instructions-data separation.
- [EXTERNAL_DOWNLOADS]: The skill references and optionally fetches external UI design materials and journey examples from the author's platform (uxkin.com). These resources provide visual and structural patterns to inform the design process.
Audit Metadata