prototype-to-production
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill workflow involves ingesting and processing untrusted design files which serves as a potential attack surface.\n
- Ingestion points: The agent reads HTML files and Figma exports from the project workspace, including paths like
.superdesign/design_iterations/*.html.\n - Boundary markers: The instructions do not define specific delimiters or ignore instructions to prevent the agent from following directions potentially embedded within the prototype files.\n
- Capability inventory: The skill performs design analysis and generates TypeScript React components, creating a path from external data to code generation.\n
- Sanitization: There are no explicit steps provided for sanitizing or escaping content extracted from prototypes before it is processed by the code generation logic.\n- [SAFE]: No other security findings were detected. The skill instructions and templates adhere to standard software development practices, using well-known frameworks and industry-standard schemas for design tokens.
Audit Metadata