literature-survey

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

• Third-party content exposure detected (low risk: 0.10). The literature-survey skill downloads research papers, web pages, repositories, and forum/blog content via automated scripts (fetch_artifacts.py), where outsider-authored free text from external sources is ingested and read by agents.

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

• Potentially malicious external URL detected (medium risk: 0.30). The skill queries arXiv, Semantic Scholar, OpenAlex, Crossref, and GitHub during setup and runtime via its scripts, which act as strong official and reputable third-party operational dependencies.