supabase-ts

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

• Third-party content exposure detected (high risk: 0.80). This skill clearly ingests untrusted, user-provided content via the Edge Function (supabase/functions/embed/index.ts) and stores it in public.documents (references/ai-vectors.md) and then exposes that content back through semantic-search RPCs (public.match_documents / match_documents_filtered) which the application/agent is expected to read/interpret as part of its workflow, creating a clear vector for indirect prompt injection.