fetching-blocked-urls

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

• Third-party content exposure detected (high risk: 0.90). This skill's SKILL.md explicitly instructs the agent to fetch and render arbitrary public web pages via Jina's reader (see the Core Command curl "https://r.jina.ai/TARGET_URL" and the Workflow Integration fallback), meaning untrusted third‑party content is ingested and could materially influence agent behavior and enable indirect prompt injection.

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

• Potentially malicious external URL detected (high risk: 0.80). The skill explicitly runs curl against https://r.jina.ai/TARGET_URL at runtime to fetch markdown that is injected as the page content/fallback into the agent workflow, so remote content from r.jina.ai can directly control prompts.