jina-reader

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

• Third-party content exposure detected (high risk: 1.00). The skill's scripts (scripts/reader.sh) fetch and return arbitrary public web content and search results via Jina endpoints (e.g., https://r.jina.ai/${INPUT} and https://s.jina.ai/${ENCODED_QUERY}), causing the agent to ingest untrusted, user-provided third‑party webpages/search results as part of its workflow.