pixel

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

• Third-party content exposure detected (high risk: 0.80). The workflow explicitly instructs the agent to call the Figma MCP (get_design_context and get_screenshot) to ingest Figma node content — arbitrary user-generated third-party data — which the agent reads and uses to map UI and decide how to generate code, allowing untrusted content to influence actions.