deepgram-rust-speech-to-text

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

• Third-party content exposure detected (medium risk: 0.65). The skill’s runtime path ingests outsider-authored free text when the operating user transcribes URLs or files they supply (e.g., transcription().prerecorded(source, ...) / .file(...)), because the resulting transcript text is derived from that external audio content and then fed back into the agent/LLM context as readable transcript output.