semantic-scholar

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

• Third-party content exposure detected (high risk: 0.75). This skill’s required workflow calls Semantic Scholar (via tools/semantic_scholar_fetch.py or inline urllib to https://api.semanticscholar.org/...) and then ingests returned paper fields like abstract/tldr into the agent’s output context, which is outsider-authored free text from a public API response.