prototype
Pass
Audited by Gen Agent Trust Hub on Sep 16, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill requires the agent to ground its prototypes by inspecting the current product, repository, and design system. This ingestion of external data without explicit isolation measures creates a surface where malicious content in the repository could influence agent behavior.
- Ingestion points: Step 2 (Ground the shared truth) directs the agent to read repository content, components, and real data shapes.
- Boundary markers: The skill does not provide specific delimiters or 'ignore embedded instructions' warnings for the data it processes.
- Capability inventory: The process relies on tools to build, run, render, and inspect generated code.
- Sanitization: There are no requirements for sanitizing or validating external content before it is processed.
- [DYNAMIC_EXECUTION]: The skill centers on generating, building, and running disposable UI code alternatives. While this is performed in isolated workspaces and is the core purpose of the skill, it represents the dynamic execution of generated content based on external inputs.
Audit Metadata