figma-implement-design

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

• Third-party content exposure detected (medium risk: 0.30). In SKILL.md “Required Workflow” Step 2 and Step 3, the agent calls get_design_context and get_screenshot based on a user-provided Figma URL/node ID, thereby ingesting design context and screenshot/asset payloads originating from the referenced (outsider-authored) Figma content.

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

• Potentially malicious external URL detected (high risk: 0.80). The skill requires parsing and using user-provided Figma design URLs (e.g., https://figma.com/design/kL9xQn2VwM8pYrTb4ZcHjF/DesignSystem?node-id=42-15 and https://figma.com/design/pR8mNv5KqXzGwY2JtCfL4D/Dashboard?node-id=10-5) at runtime to call get_design_context/get_screenshot, so the fetched design data from those URLs directly controls the agent's generation behavior.