figma-mcp-integration

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

• Third-party content exposure detected (low risk: 0.10). The skill integrates with Figma via Model Context Protocol (MCP) to read design structures, node info, and text content. Since Figma designs are authored by users and collaborators within an organization or public workspace, reading them involves ingesting user-generated content, but design files are typically considered own-tenant or trusted collaborative workspace data rather than unmonitored external feeds, or fall under low/no-risk user file access depending on scoping.

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

• Potentially malicious external URL detected (medium risk: 0.30). The skill instructs cloning an unofficial GitHub repository (https://github.com/grab/cursor-talk-to-figma-mcp.git) and installing an unofficial package (cursor-talk-to-figma-mcp@latest) via setup instructions, constituting a common/reputable unofficial operational dependency.