figma-ds-rules

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

• Third-party content exposure detected (high risk: 0.75). The required workflow explicitly says to “Load Figma's current design-system-rules skill from the official MCP guide distribution” (a public web/GitHub source), which at runtime would ingest outsider-authored free text into the agent’s LLM context.

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

• Potentially malicious external URL detected (high risk: 0.90). SKILL.md explicitly instructs the skill to "Load Figma's current design-system-rules skill from the official MCP guide distribution", and the official source URL https://github.com/figma/mcp-server-guide is listed and would be fetched at runtime to control the agent's rules/prompts.