user-research-cookiy

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

• Third-party content exposure detected (high risk: 0.75). The required workflow for Route B/C ultimately ingests outsider-authored participant transcripts/notes (from Cookiy study interview playback/content and report content) as readable text into the synthesis sub-agents’ LLM context for coding/summarization, creating an indirect prompt-injection risk.