show-me-the-design-cn

Pass

Audited by Gen Agent Trust Hub on Nov 1, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests and analyzes user-provided source code as its primary function, which creates an attack surface for instructions embedded within the source data (e.g., in code comments).
  • Ingestion points: User-supplied source code files and project documentation (referenced in SKILL.md).
  • Boundary markers: Instructions in SKILL.md and references/views.md to verify design against source evidence and distinguish between documented implementations and inferred designs.
  • Capability inventory: Static HTML/JS/CSS generation for documentation delivery.
  • Sanitization: No explicit instructions are provided to sanitize or escape user-controlled strings before they are interpolated into the generated HTML report templates.
  • [DYNAMIC_EXECUTION]: The skill generates interactive HTML documents containing CSS and JavaScript. While the provided template in 'references/design-system.md' focuses on benign UI interactions like scroll tracking and navigation, the agent's assembly of executable web code from analyzed external content constitutes a form of dynamic script generation.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 1, 2026, 11:23 AM