penpot-migrate

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

• Third-party content exposure detected (medium risk: 0.65). The required workflow ingests Figma source data (via the Figma MCP design-context tools or via user-pasted Figma export/JSON) and then normalizes it into the IR that is subsequently used by execute_code, so outsider-authored free text/JSON from Figma can become LLM-readable context through the “pasted data”/Figma MCP inventory path.