fe-design-code

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 1.00). The skill explicitly fetches and ingests untrusted, user-generated Figma content via the Figma API/MCP (see "1. Resolve the target Figma node" and numerous references to MCP/REST outputs) and then reads and interprets that content to generate code, choose CSS/viewport, and drive verification/VRT, so third-party content can directly influence agent actions.