The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes the belt command-line interface to perform audio generation tasks. This tool usage is explicitly scoped within the allowed-tools configuration in the frontmatter, limiting the agent's shell access to only the necessary utility.
[EXTERNAL_DOWNLOADS]: The documentation references installation scripts and assets hosted on the inference.sh domain and its official GitHub repositories. These are standard resources for the platform's infrastructure.
[PROMPT_INJECTION]: The skill handles user-provided text descriptions which are passed as arguments to the sound generation tool. This is a characteristic of the intended text-to-audio functionality.
Ingestion points: User-provided text strings are used in the text parameter within shell commands.
Boundary markers: The text input is encapsulated within a structured JSON object passed to the CLI command arguments.
Capability inventory: The skill is restricted to executing the belt CLI tool for application runs.
Sanitization: No explicit sanitization of user input is documented within the instruction file, relying on the underlying API's handling.

elevenlabs-sound-effects