figma-implement-design

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

• Third-party content exposure detected (medium risk: 0.30). In the required workflow, outsider-provided Figma URLs (file URL/node ID supplied by the user) are parsed and then the agent calls the Figma MCP to fetch design context and screenshots/assets for that node, meaning attacker-authored free text from the referenced Figma (e.g., text layers/copy) is ingested via MCP tools without selecting a specific trusted item first.

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

• Potentially malicious external URL detected (high risk: 0.70). The skill explicitly requires calling a Figma MCP at runtime to fetch design data that directly drives the agent's implementation instructions (e.g., the example URL https://figma.com/design/:fileKey/:fileName?node-id=1-2), so external content would control the agent's outputs.