figma-import

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

• Third-party content exposure detected (medium risk: 0.30). Source “figma.com link” content is ingested via the required runtime workflow calling Figma reading functions like get_metadata, get_design_context, get_variable_defs, and get_screenshot, where user-provided Figma links can reference outsider-authored design text/layer names/tokens.

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

• Potentially malicious external URL detected (high risk: 0.80). The skill explicitly requires using a figma.com link at runtime (figma.com/design/:fileKey/:name?node-id=1-2) and to call runtime pulls like get_design_context/get_metadata/get_variable_defs that fetch Figma content which directly guides the agent's extraction and instructions, making the external URL a required runtime dependency that controls the agent's behavior.