speech-to-text

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

• Third-party content exposure detected (high risk: 0.90). The skill's SKILL.md shows the agent accepting arbitrary external media via the top-level "image_urls" parameter (e.g., MP3/MP4 URLs like "https://example.com/...") and explicitly transcribes those public audio/video files, which are untrusted third‑party content the agent reads and whose spoken instructions could influence outputs and downstream actions.