tavily-literature-review

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

• Third-party content exposure detected (low risk: 0.10). The workflow searches external academic indices and databases (such as arXiv, PubMed, SSRN, and journals) via Tavily and reads individual research papers and preprints, which constitutes reading user/outsider-searchable web content rather than monitoring an unselected outsider submission feed.

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

• Potentially malicious external URL detected (medium risk: 0.30). The skill references npx skills add tavily-ai/skills --skill tavily-search, which fetches code from an external repository during setup.