narrative-resonance-monitor

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

• Third-party content exposure detected (medium risk: 0.30). At runtime, this skill probes outsider-authored public content via scripts/connectors/tavily.py --answer, scripts/connectors/bluesky.py, and scripts/connectors/gdelt.py to derive AI-answer perception/news/social resonance signals (even though labeled proxy), so attackers can post poison into those sources and have it ingested without the agent first selecting a specific trusted item.