voice-ai
Warn
Audited by Snyk on Feb 15, 2026
Risk Level: MEDIUM
Full Analysis
MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).
- Third-party content exposure detected (high risk: 0.90). This skill ingests untrusted third-party caller audio via Twilio Media Streams (websocket), transcribes it with Deepgram, and feeds the resulting user-generated text into the Groq LLM for interpretation, which clearly allows arbitrary callers to inject instructions indirectly.
Audit Metadata