full-page-translate

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

• Third-party content exposure detected (medium risk: 0.30). Source: the required workflow reads arbitrary outsider-authored free text by fetching the full article from a user-supplied URL via GET markdown content from r.jina.ai/{url} and then analyzes/translates the entire retrieved content.

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

• Potentially malicious external URL detected (high risk: 0.80). The skill explicitly GETs markdown at runtime from r.jina.ai (e.g., "https://r.jina.ai/https://example.com/article" and "r.jina.ai/{url}") and injects that fetched content into the model for translation, which is equivalent to fetching remote text that can directly influence prompts/behavior.