pixel

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

• Third-party content exposure detected (high risk: 1.00). The workflow explicitly instructs the agent to call the Figma MCP (see "Workflow -> Analyze the Request" and "MCP Usage -> Figma MCP") using get_design_context and get_screenshot to ingest user Figma nodes/designs (untrusted, user-generated third-party content) which the agent must read and use to drive code generation, creating a clear avenue for indirect prompt injection.