entity-optimizer

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

• Third-party content exposure detected (high risk: 0.70). The required workflow includes runtime calls to external public sources (e.g., scripts/connectors/pageviews.py and scripts/connectors/gdelt.py for Wikipedia article titles and GDELT news mentions), which can ingest outsider-authored free text (news/article content) into the agent’s LLM context for summarization/evidence.