product-page-design

Pass

Audited by Gen Agent Trust Hub on Mar 16, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill handles untrusted data from product descriptions and reviews when generating React components and JSON-LD structured data, creating an indirect prompt injection surface. * Ingestion points: Data enters via the buildProductSchema function and component props in SKILL.md and the task files. * Boundary markers: Absent; dynamic content is interpolated directly into the UI and schema templates without delimiters. * Capability inventory: The skill's capabilities are limited to UI rendering and metadata generation; it lacks access to high-risk APIs such as file system writing or remote command execution. * Sanitization: No specific sanitization or filtering logic is provided in the code samples to process external strings.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 16, 2026, 12:39 PM