design-prototype
Pass
Audited by Gen Agent Trust Hub on Oct 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to generate code (HTML demos and React components) based on user prompts and existing codebase context. This creates a surface where an attacker could provide malicious design questions or data models that contain embedded instructions intended to influence the agent's output or the environment in which the prototype is run.
- Ingestion points: User prompts, surrounding code, and project artifacts identified through project-specific tools (referenced in
SKILL.md). - Boundary markers: The instructions do not specify the use of delimiters or 'ignore embedded instructions' warnings when processing user requirements for the prototypes.
- Capability inventory: The skill involves writing files to the local file system (HTML, TSX) and provides commands for the user/agent to run project tasks using
pnpm,python, orbun(SKILL.md). - Sanitization: The instructions do not mandate escaping or validating user-provided strings before they are interpolated into the generated prototype source code.
Audit Metadata