analyze-video

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

• Third-party content exposure detected (high risk: 0.85). The skill’s runtime path ingests outsider-authored free text via Step 2: audio transcription—it runs Whisper on the extracted audio.wav from the user-provided reference video, producing a readable metadata.txt/transcript text that is then used by the agent to generate the template (the spoken dialogue is not authored by the operating user).