product-design-design-qa

Pass

Audited by Gen Agent Trust Hub on Jul 10, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes a local bootstrap script (bootstrap-prototype.mjs) found within the plugin directory to initialize project structures for UI prototypes. This is a standard functional component for the tool's intended purpose.
  • [EXTERNAL_DOWNLOADS]: The skill utilizes npm install to manage project dependencies during the prototype creation phase. This involves fetching packages from the official registry, which is an expected part of the development lifecycle.
  • [PROMPT_INJECTION]: The skill processes content from user-provided URLs and design platforms like Figma. This constitutes a surface for indirect prompt injection. However, as this is the primary functionality of a QA tool (reading and analyzing external implementations), it is considered an acceptable and expected risk.
  • Ingestion points: External implementation URLs, local previews, and Figma design nodes accessed during the QA workflow.
  • Boundary markers: No explicit instructions were found to delimit external content or warn the model about potential embedded instructions within implementation code.
  • Capability inventory: The skill has the ability to write reports to the file system (design-qa.md) and execute shell commands (node, npm).
  • Sanitization: The instructions do not specify any sanitization or filtering of external implementation content before it is processed for comparison.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 10, 2026, 06:05 PM
Security Audit — agent-trust-hub — product-design-design-qa