ai-music-generation
Pass
Audited by Gen Agent Trust Hub on Nov 14, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill references external installation resources and additional utility skills necessary for its music generation functionality. \n
- Links to CLI installation instructions hosted on GitHub:
https://raw.githubusercontent.com/inference-sh/skills/refs/heads/main/cli-install.md. \n - Recommends the addition of the
belt-sh/cliskill to the agent's environment usingnpx. \n- [COMMAND_EXECUTION]: The skill guides the agent to use shell commands to interact with the music generation service. \n - Instructs the use of
belt loginandbelt app runfor music generation tasks. \n - The
allowed-toolsconfiguration correctly limits the agent's Bash usage to commands prefixed withbelt. \n- [INDIRECT_PROMPT_INJECTION]: The skill processes user-generated text for music prompts and lyrics, which are passed as inputs to external commands. \n - Ingestion points: User-supplied prompts and lyrics are used as inputs for the
belt app runcommand inSKILL.md. \n - Boundary markers: No explicit delimiters or instructions are provided to the agent to treat music prompts as data and ignore embedded instructions. \n
- Capability inventory: The skill utilizes the
Bashtool to perform its primary music generation functions. \n - Sanitization: The documentation does not describe explicit sanitization or escaping of the user-provided strings before they are interpolated into the JSON structure of the shell command.
Audit Metadata