figma-design-systems

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

• Third-party content exposure detected (high risk: 0.75). The required workflow instructs the agent to open Figma’s official MCP guide and developer documentation (public web content) at runtime, which can include outsider-authored free text that the LLM reads for verification/citation.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.90). The SKILL explicitly instructs the agent at runtime to "Open Figma's current official MCP guide and developer documentation," meaning the agent will fetch and use https://github.com/figma/mcp-server-guide and https://developers.figma.com/ to inform and control its responses.