academic-research-skills-codex

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 perform literature searches and syntheses from public third-party sources (see "Parallel Literature Search" and the "deep-research"/"literature review" workflows referencing Google Scholar, Scopus, Web of Science, ERIC/ProQuest and "grey literature"), so the agent is expected to fetch and interpret untrusted public content that can materially influence its actions.