designmd-supply-generator

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

• Third-party content exposure detected (high risk: 0.95). The required runtime workflow fetches outsider-authored free text from arbitrary public websites via client.web.webScrapeMd({ url: https://${domain} }) (and also styleguide/brand/screenshot-derived text) and then injects it into the LLM prompt as Page content: ${markdown.content} before calling generateText.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill fetches arbitrary webpage content and screenshots at runtime (e.g., via client.web.webScrapeMd / client.web.screenshot using URLs like https://${domain}) and injects that remote content (page markdown and images) directly into the LLM prompt, so external sites can directly control the generated instructions.