moai-design-tools

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

• Third-party content exposure detected (high risk: 1.00). The skill explicitly instructs the agent to fetch and ingest arbitrary public web pages and user-generated Figma/FigJam content (see reference/figma.md: generate_figma_design(url, ...) and get_figjam via the remote MCP server https://mcp.figma.com/mcp), and those external contents are read and used to drive design generation and code-export workflows, exposing the agent to untrusted third-party input that could carry indirect prompt injection.