voice-analyzer

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

• Third-party content exposure detected (high risk: 0.80). The skill's Phase 1 "Gather Samples" explicitly requires the agent to ingest 3–5 writing samples (published or user-provided text) which the agent will read and analyze, exposing it to arbitrary third‑party/untrusted content that could contain indirect prompt injections.