frontend-design
Pass
Audited by Gen Agent Trust Hub on Feb 21, 2026
Risk Level: SAFENO_CODE
Full Analysis
- Prompt Injection (SAFE): Natural instructional language such as 'CRITICAL' and 'IMPORTANT' is used to define stylistic goals and does not attempt to bypass agent safety filters or override system instructions.
- NO_CODE (SAFE): The skill does not package any executable scripts, binaries, or configuration files.
- Indirect Prompt Injection (SAFE): A surface for indirect injection is present: 1. Ingestion points: User-provided frontend requirements in the prompt context. 2. Boundary markers: Absent. 3. Capability inventory: Generates functional frontend code (HTML, JS, React, Vue). 4. Sanitization: Absent.
Audit Metadata