literature-review

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

• Third-party content exposure detected (high risk: 0.75). SKILL.md’s required workflow includes fetching/aggregating literature results (e.g., via gget/arXiv/Semantic Scholar/Google Scholar) and then formatting those results into markdown (scripts/search_databases.py), which can include outsider-authored free text like titles/abstracts from public databases into the agent’s LLM context.