create-design-system-rules

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

• Third-party content exposure detected (high risk: 0.80). The skill's required workflow explicitly calls the Figma MCP server (steps requiring get_design_context, get_screenshot, and the assets endpoint) and instructs the agent to download and act on Figma-provided content (including following asset URLs such as localhost), meaning user-generated/untrusted Figma payloads are fetched and directly influence implementation decisions.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill requires runtime calls to a Figma MCP server (see https://developers.figma.com/docs/figma-mcp-server/)—its create_design_system_rules/get_design_context responses are fetched during execution and return the foundational prompt/templates that directly control the agent's instructions and are required for the skill to operate.