elevenlabs-dubbing
Pass
Audited by Gen Agent Trust Hub on Jul 7, 2026
Risk Level: SAFE
Full Analysis
- [COMMAND_EXECUTION]: The skill uses the
Bashtool to interact with thebeltCLI. Access is restricted via theallowed-tools: Bash(belt *)frontmatter, which limits the agent to only executing commands starting with 'belt', following the principle of least privilege. - [EXTERNAL_DOWNLOADS]: The skill references installation scripts and additional resources from the official
inference-shGitHub repositories. These references are used for setup and expanding functionality within the intended service ecosystem and do not point to untrusted or suspicious sources. - [CREDENTIALS_SAFE]: User authentication is managed through the
belt logincommand. This is a secure approach that avoids hardcoding API keys or secrets within the skill instructions or script files.
Audit Metadata