design-taste-frontend-v1
Pass
Audited by Gen Agent Trust Hub on Oct 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is configured to dynamically adapt design variables based on user chat prompts, creating a surface where external input influences the agent's output logic.
- Ingestion points: The agent is instructed to listen to user chat prompts to override the ACTIVE BASELINE CONFIGURATION values (DESIGN_VARIANCE, MOTION_INTENSITY, VISUAL_DENSITY) defined in SKILL.md.
- Boundary markers: There are no specific delimiters or guardrail instructions defined to prevent the agent from obeying potentially malicious instructions embedded in the user's styling requests.
- Capability inventory: The agent generates frontend code and provides terminal commands for package installations (npm install).
- Sanitization: The skill does not implement input validation or sanitization for the requested design overrides.
- [EXTERNAL_DOWNLOADS]: The skill suggests using external assets for visual components.
- Evidence: Recommends using
https://picsum.photos/for reliable image placeholders instead of other services.
Audit Metadata