deepclaw-voice

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

• Third-party content exposure detected (medium risk: 0.30). The required workflow sets up a voice agent server (Deepgram Voice Agent API + Twilio) that accepts inbound phone calls and processes speech/text through an LLM proxy, allowing callers to submit arbitrary free text.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.70). The skill instructs cloning an external repository from GitHub (https://github.com/deepgram/deepclaw.git) and installing its requirements at runtime, creating an unverified runtime external dependency.