design-to-wp-block

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

• Third-party content exposure detected (high risk: 1.00). The skill explicitly fetches and ingests designs from Figma (see Tier 2 "Fetch Design Data in Main Conversation" using mcp__plugin_figma_figma__get_design_context and mcp__plugin_figma_figma__get_screenshot), which are untrusted/third-party user-generated content that the agent must read and use to drive generation and downstream tool actions.

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

• Potentially malicious external URL detected (high risk: 1.00). The skill fetches user-provided Figma file URLs (e.g., https://www.figma.com/file/...) at runtime via the mcp__plugin_figma_* calls and injects the extracted design specifications directly into the Task(prompt) that controls the spawned design-specialist, so the external Figma content directly controls agent instructions.