using-jina-reader

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

• Third-party content exposure detected (high risk: 1.00). The SKILL.md explicitly instructs the agent to fetch and read arbitrary public web pages via Jina Reader (https://r.jina.ai/) using WebFetch, which exposes untrusted third‑party content that the agent will interpret and could materially influence its actions.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill instructs the agent to call https://r.jina.ai/ at runtime to fetch and inject arbitrary markdown page content into the model context (used as a fallback for blocked WebFetch), so the external URL can directly control prompts.