creative-research

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

• Third-party content exposure detected (high risk: 0.85). The skill’s workflow is to “search for current sources” and gather “wikis and encyclopedias,” “community discussion,” and “social media and interviews,” which implies runtime ingestion of outsider-authored free text from public web pages/forums/social posts into the agent’s LLM context.