literature-review

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

• Third-party content exposure detected (low risk: 0.10). The required workflow ingests and screens search results from Semantic Scholar and arXiv containing outsider-authored paper abstracts, titles, and metadata.

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

• Potentially malicious external URL detected (medium risk: 0.30). The optional arxiv-search skill is documented with an external repository URL (https://github.com/dsebastien/ai-skill-arxiv) for installation, representing a third-party unofficial operational dependency.