prototype
Pass
Audited by Gen Agent Trust Hub on Oct 5, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied UI descriptions to generate functional code variants. This creates a potential attack surface where instructions embedded in the description could attempt to influence the generated code or the final integration into production files.\n
- Ingestion points: User-provided
<description>input inSKILL.mdPhase 1.\n - Boundary markers: The instructions do not define explicit delimiters or instructions to ignore potential commands within the user's description.\n
- Capability inventory: The skill performs file system operations, including creating prototype files (Phase 4) and modifying production code (Phase 6).\n
- Sanitization: There is no specified logic for sanitizing or validating the input description before it is interpolated into code generation tasks.\n- [DYNAMIC_EXECUTION]: The prototype harness specified in
PICKER.mdutilizes dynamic JavaScript execution techniques, such as usinginnerHTMLto swap between generated UI components andrequestAnimationFramefor rendering control. These are standard practices for lightweight UI prototyping and local preview environments.
Audit Metadata