implementing-figma

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

• Third-party content exposure detected (high risk: 0.90). The skill explicitly instructs the agent to fetch and parse public Figma URLs and shared Make links (Step 1 Option A; §2 get_design_context, get_metadata, get_screenshot; §4 Figma Make integration), which ingest user-generated/untrusted third‑party content and then directly drive implementation decisions and tool use.

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

• Potentially malicious external URL detected (high risk: 1.00). The skill requires contacting the remote MCP endpoint https://mcp.figma.com/mcp at runtime to fetch design artifacts and generated design-system-rules (and other outputs) which are then injected into prompts/instructions (e.g., "以下のデザインシステムルールに従って実装してください: [rules]"), so external content directly controls agent instructions and is a required dependency.