academic-research

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

• Third-party content exposure detected (high risk: 0.80). The skill's search_papers.py explicitly queries the public Semantic Scholar API (https://api.semanticscholar.org/graph/v1/paper/search) and ingests fields like abstracts and Semantic Scholar "tldr" which the agent reads and uses for sorting/selection, so it consumes untrusted third-party content that can influence decisions.