elevenlabs-music

Pass

Audited by Gen Agent Trust Hub on Aug 28, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill references documentation and skill packages from the official GitHub repositories of the service provider (inference-sh). These are well-known sources within the context of the skill's functionality.
  • [COMMAND_EXECUTION]: The skill utilizes a command-line interface tool (belt) to authenticate and interact with the music generation API. The execution environment is appropriately scoped to this specific tool using platform restrictions.
  • [INDIRECT_PROMPT_INJECTION]: The skill takes user-provided text prompts to generate music. This input is interpolated into shell commands for the CLI tool. While this represents a standard injection surface for AI models, the scope is limited to audio content generation and does not pose a risk to the underlying system.
  • Ingestion points: The prompt field in the JSON input for the belt command (SKILL.md).
  • Boundary markers: None present in the command examples.
  • Capability inventory: Uses Bash(belt *) to execute platform-specific commands.
  • Sanitization: No specific sanitization or filtering of prompt content is described.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 28, 2026, 01:13 PM
Security Audit — agent-trust-hub — elevenlabs-music