elevenlabs-voice-isolator
Pass
Audited by Gen Agent Trust Hub on Aug 31, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill references the installation of the
belt-sh/clitool and other related skills from theinference-shorganization. These are well-known resources associated with the inference.sh platform for which this skill is designed. - [INDIRECT_PROMPT_INJECTION]: The skill interacts with external audio data by passing URLs to the
beltCLI tool. - Ingestion points: Audio URLs are passed as arguments to the
belt app runcommand inSKILL.mdexamples. - Boundary markers: The skill uses JSON objects to structure the input arguments, reducing the risk of argument injection.
- Capability inventory: The skill is restricted to running the
beltcommand viaBash(belt *)in theallowed-toolsconfiguration. - Sanitization: Not applicable as the skill primarily provides documentation for CLI usage rather than complex script logic.
- [SAFE]: No malicious patterns such as obfuscation, credential exfiltration, or unauthorized command execution were detected. The skill's configuration limits the agent's actions to a specific, legitimate tool.
Audit Metadata