augment
Pass
Audited by Gen Agent Trust Hub on Aug 11, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [COMMAND_EXECUTION]: The skill uses the
bashtool to execute local maintenance scripts (kokoro-install.sh) and verification commands to manage the TTS environment. - [EXTERNAL_DOWNLOADS]: It automates the installation of standard Python packages (
mlx-audio,soundfile,numpy) and fetches model weights from themlx-communityrepository on Hugging Face, which is a well-known service for machine learning models. - [SAFE]: The skill includes instructions for 'self-evolution,' directing the agent to refine its own documentation if errors are found during execution. This represents a functional design choice for maintaining instruction accuracy and does not involve malicious persistence or privilege escalation.
Audit Metadata