omnivoice

Pass

Audited by Gen Agent Trust Hub on Aug 14, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses curl commands to interact with a local API endpoint at http://localhost:3900 for health checks, voice discovery, speech synthesis, and audio transcription. These operations are limited to the local network and are consistent with the skill's stated purpose of providing offline voice services.
  • [EXTERNAL_DOWNLOADS]: The skill references the standard openai Python SDK as a method for interacting with the local backend. It does not perform any unverified remote code execution or download untrusted scripts.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 14, 2026, 07:14 AM
Security Audit — agent-trust-hub — omnivoice