design-ui
Pass
Audited by Gen Agent Trust Hub on Sep 23, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill's architecture includes a 'design-researcher' agent responsible for gathering context through web research and competitor studies. This externally sourced, untrusted data is integrated into the system's context and used to drive the 'design-generator' and 'design-reviewer' agents. This creates a vulnerability where malicious instructions embedded in research targets could influence the agent's output or behavior.
- Ingestion points: External research findings, competitor analysis reports, and industry trend data are ingested into the agent context (as described in
references/agent-quality-gates.md). - Boundary markers: The provided instructions do not specify any delimiters, boundary markers, or 'ignore' directives to prevent the agent from following instructions embedded within the researched data.
- Capability inventory: The system is capable of production-level file creation (HTML, CSS, JSON), state management via local files, and multi-agent coordination with high autonomy.
- Sanitization: There is no evidence of data sanitization, validation, or filtering of the research findings before they are utilized in the generation and refinement phases.
Audit Metadata