ai-voice-cloning
Pass
Audited by Gen Agent Trust Hub on Jul 7, 2026
Risk Level: SAFE
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill references external installation instructions and related skill packages hosted on GitHub repositories associated with the inference.sh service.
- [COMMAND_EXECUTION]: Functional operations are performed through the 'belt' CLI. The execution environment is restricted to this specific tool through the allowed-tools configuration in the frontmatter.
- [PROMPT_INJECTION]: The skill documentation includes templates that interpolate user-provided content into shell command arguments. This represents a potential surface for indirect prompt injection or schema confusion, though it is mitigated by the restricted scope of the permitted CLI tool.
- Ingestion points: prompt and text variables in SKILL.md command examples
- Boundary markers: Absent
- Capability inventory: Shell command execution restricted to the 'belt' tool
- Sanitization: Not specified in the provided instruction templates
Audit Metadata