ocr-and-documents

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

• Third-party content exposure detected (high risk: 0.90). The SKILL.md explicitly instructs the agent to fetch and process remote, public URLs via web_extract (e.g., "always try web_extract first" with examples like https://arxiv.org/pdf/...), which are untrusted third‑party sources that the agent will read and act on as part of its workflow, so fetched content could influence subsequent processing or decisions.