paper-to-agent-guide

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

• Third-party content exposure detected (high risk: 0.70). The skill explicitly instructs the agent to fetch and parse published research papers and external databases (see "Paper Parsing Workflow" and "Integration with Research-Claw" — "Use literature search skills to find papers for processing" and "Link to external knowledge bases (OpenAlex, CrossRef, DOI)"), so it ingests third‑party public content that the agent reads and uses to drive decisions and actions.