share-of-voice-tracker

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

• Third-party content exposure detected (high risk: 0.75). The skill ingests outsider-authored free text via runtime-fetched public web/news content (e.g., gdelt.py news echo and tavily.py web chatter) where article/post body text is treated as untrusted input and can be included in the LLM context for labeling/proxy counts.