discovery-research-weapon

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

• Third-party content exposure detected (high risk: 0.85). The outsider-free-text injection path is via the runtime incorporation of customer interview transcripts/notes into generated artifacts (e.g., interview-script templates → OST nodes), where the participant quotes are authored by non-operating-user outsiders and can contain prompt-injection strings.