grounded-citations

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

• Third-party content exposure detected (low risk: 0.10). Although the skill supports “chat answers” and expects retrieval outputs to be registered via scripts/sources.py ingest/add, the required workflow itself does not define an outsider-authored text ingestion path (it only assigns citations to URLs provided by retrieval outputs), and evidence attachment requires verbatim text from the fetched page that the agent chose at retrieval time.