analytics-stanley-druckenmiller

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

• Third-party content exposure detected (high risk: 0.85). The skill’s runtime workflow loads references/05-current-views.md, whose “Source” items include third-party web content (e.g., CNBC transcripts/summaries and RealClearPolitics video pages) that are outsider-authored and then ingested as readable text into the agent’s LLM context via the reference-loading step.