figma-to-code-skill

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

• Third-party content exposure detected (high risk: 0.85). The skill’s runtime workflow (Step 1) explicitly reads/extracts design tokens and component details from a Figma file via Figma MCP; that Figma file content is authored by designers other than the operating user, so it can include free-form text (e.g., labels, annotations) that the LLM ingests as readable prose.