youmind-x-article

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

• Third-party content exposure detected (high risk: 0.75). The skill’s runtime pipeline (Step 2 “Mine YouMind knowledge base for relevant material” and Step 3 “Research topic via web search”) calls YouMind APIs that return outsider-authored web/page/article text, which is then used as LLM context for content generation/adaptation before publishing.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill makes runtime calls to the YouMind OpenAPI (e.g., https://youmind.com/openapi/v1) — including createChat/listMessages in agent mode that load remote "generateImage" tools and drive publishing — so it relies on remote API responses that execute agent tools and directly control the content/prompting flow.