message-system-architect

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

• Third-party content exposure detected (medium risk: 0.30). This skill’s runtime workflow reads positioning truth set content from memory/narrative/positioning-truth-tracer/ (or pasted) and the change-narrative arc from memory/narrative/strategic-narrative-designer/, plus “personas (User-provided)”, meaning an outsider can inject free text via the user-provided/pasted upstream inputs that the LLM ingests directly without selecting a specific item.