extract-design-system

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

• Third-party content exposure detected (high risk: 0.85). The required workflow includes fetching a user-supplied public website URL at runtime via npx extract-design-system <url>, and the resulting extracted .extract-design-system/normalized.json/assets are then read and summarized by the agent—i.e., outsider-authored free text from public web content can flow into the LLM context.