clipping

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

• Third-party content exposure detected (medium risk: 0.30). /clipping reads and LLM-ingests transcript text and transcript chunks from the Supabase long-form vault (library.longform_assets.transcript and library.transcript_chunks.text) to perform semantic search and full transcript reads, and those vault texts ultimately originate from outsider-authored recording/transcript content.