biorxiv-database

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

• Third-party content exposure detected (high risk: 0.75). SKILL.md describes runtime HTTP calls to the public bioRxiv/medRxiv REST APIs (e.g., /details/...), whose JSON fields like title and abstract are outsider-authored free text that would be ingested into the agent’s LLM context if the workflow passes those strings along.