voice-ai

Pass

Audited by Gen Agent Trust Hub on Mar 22, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill serves as a technical reference and template for integrating voice-processing services. It contains no malicious intent or harmful instructions.
  • [CREDENTIALS_UNSAFE]: No hardcoded API keys or secrets were found. The skill follows the security best practice of retrieving sensitive credentials from environment variables (e.g., DEEPGRAM_API_KEY, GROQ_API_KEY, CARTESIA_API_KEY).
  • [EXTERNAL_DOWNLOADS]: The skill references standard Python libraries (deepgram-sdk, groq, cartesia, twilio, fastapi, httpx) from official package registries. These are necessary and appropriate for the skill's stated purpose.
  • [DATA_EXFILTRATION]: No unauthorized data transmission patterns were identified. Network operations are limited to established communication with the specified voice and AI service providers.
  • [PROMPT_INJECTION]: While the skill is designed to process untrusted user audio input, it includes explicit, "non-negotiable" guardrails and prompt engineering templates to prevent manipulation of the agent's behavior. These measures include strict rules on information disclosure and automated escalation triggers for human intervention.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 22, 2026, 06:50 PM