audio-transcriber

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

• Third-party content exposure detected (high risk: 0.90). The skill explicitly accepts and downloads audio from arbitrary URLs (SKILL.md Step 1: "Accept file path or URL: https://example.com/audio.mp3 (download to temp directory)") and then transcribes and feeds the resulting transcript into prompt-engineer/LLM pipelines and extraction agents to generate action items/decisions, so untrusted third-party audio can be converted to text that materially influences prompts, outputs, and downstream processing.