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 arbitrary audio from URLs (SKILL.md Step 1: "Accept file path or URL" — download to temp directory) and its implementation (scripts/transcribe.py: handle_prompt_workflow and process_with_llm) transcribes that untrusted, third‑party audio and feeds the resulting text into prompt-engineering and LLM CLIs to generate summaries/action items that materially influence subsequent outputs and decisions.