omh-voice-input
Pass
Audited by Gen Agent Trust Hub on Sep 25, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill establishes guidance for managing terse voice and mobile commands, emphasizing clarity and user-facing status cards.
- [SAFE]: Includes explicit safety rules that prohibit the agent from claiming external platform actions or execution success without actual observed evidence.
- [SAFE]: The inclusion of Korean text (e.g., '음성 명령', '모바일 명령') serves as legitimate routing signals for multi-language voice commands and does not represent obfuscation or homoglyph attacks.
- [SAFE]: The shell command example (
omh runtime record) identifies a local tool harness for recording workflow status and does not involve remote code execution, unauthorized downloads, or privilege escalation.
Audit Metadata