figma

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

• Third-party content exposure detected (high risk: 0.75). The required runtime workflow explicitly calls Figma MCP tools like get_design_context/get_screenshot, whose responses are derived from user-provided Figma node content (outsider-authored design text/metadata) and are fed back into the agent’s LLM context.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill declares and requires the Figma MCP server at https://mcp.figma.com/mcp (used at runtime via the MCP tool) which returns design context and generated React/Tailwind code that directly drives agent implementation decisions and thus controls prompts/output.