prototype
Pass
Audited by Gen Agent Trust Hub on Oct 11, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill defines a framework for generating logic prototypes and UI variations based on user prompts and design questions, creating an inherent surface for indirect prompt injection.
- Ingestion points: User-provided design questions, prompts, and existing code context are used as the primary drivers for generating logic modules and UI components (referenced in SKILL.md, LOGIC.md, and UI.md).
- Boundary markers: The instructions do not direct the agent to implement security delimiters or "ignore instructions" blocks when interpolating user-controlled content into the generated prototype source code.
- Capability inventory: The skill instructs the agent to create interactive files (HTML/JS/TSX) on the local filesystem that can execute logic and manipulate state, providing a mechanism for potentially malicious code to be introduced via the initial prompt.
- Sanitization: The skill lacks explicit instructions for the agent to sanitize or escape external content before it is embedded into the generated prototypes.
Audit Metadata