voice-extractor

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

• Third-party content exposure detected (medium risk: 0.30). The voice-extractor skill ingests outsider-authored free text when the user provides writing samples (e.g., pasted text/file samples) for the runtime “extract” mode, and the co-installed “enforce” mode recomputes checks on the draft but still relies on whatever free text the user supplies to be measured.