telnyx-voice-media-python

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

• Third-party content exposure detected (high risk: 0.80). The skill's API and examples show it can fetch and play arbitrary audio via the audio_url parameter (Play audio URL section) and receives untrusted, user-generated content such as call.recording.transcription.saved.transcription_text and payload.media_url in webhook payloads (Webhooks / Webhook payload fields), which the agent is expected to parse and could influence subsequent actions.