designing-as
Pass
Audited by Gen Agent Trust Hub on Aug 26, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill directs the agent to research external designer portfolios (e.g., Dribbble) to extract design patterns, creating an attack surface for indirect prompt injection.
- Ingestion points: External portfolio websites (e.g., dribbble.com) accessed via browser tools (SKILL.md).
- Boundary markers: Absent; the agent is not instructed to treat external content as potentially untrusted data.
- Capability inventory: Browser usage (built-in or Playwright) and UI design generation.
- Sanitization: No sanitization or content filtering is implemented for the fetched web data.
Audit Metadata