delight
Pass
Audited by Gen Agent Trust Hub on Oct 26, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill requires the agent to ingest and analyze context from the current conversation thread and codebase to infer brand personality and audience demographics. This creates a theoretical surface where malicious instructions embedded in the analyzed data could attempt to influence the agent's behavior.
- Ingestion points: Reads the current thread history and codebase files (e.g., existing design and functionality).
- Boundary markers: No explicit boundary markers or 'ignore' instructions for external data are provided within the skill; it relies on the agent's native prompt engineering.
- Capability inventory: Executes calls to the
AskUserQuestionTool, triggers thefrontend-designskill, and generates UI code (CSS, JS) and copy. - Sanitization: There are no explicit sanitization or validation steps for the content retrieved from the thread or codebase.
- [SAFE]: The skill references established open-source projects and well-known frontend libraries (GSAP, Lottie, Framer Motion) as implementation suggestions. All identified external links point to legitimate development resources and official documentation.
Audit Metadata