prior-art-hunter

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

• Third-party content exposure detected (high risk: 0.80). The skill explicitly instructs the agent to search and read public third-party databases and publications (e.g., USPTO, Google Patents, Espacenet, WIPO PatentScope, Google Scholar, IEEE Xplore, arXiv) and ingest full patent/publication texts as part of its workflow, which exposes the agent to untrusted external content that could contain indirect prompt injection.