The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill references an installation script at https://raw.githubusercontent.com/inference-sh/skills/refs/heads/main/cli-install.md and provides instructions to install belt-sh/cli via npx. These resources are provided by the service vendor to enable the skill's core functionality.
[COMMAND_EXECUTION]: The skill demonstrates usage of the belt CLI tool for audio inference and processing. The skill's configuration (allowed-tools) restricts the agent's shell access specifically to the belt command, which follows the principle of least privilege.
[INDIRECT_PROMPT_INJECTION]: The skill accepts arbitrary text for audio generation, which presents a surface for indirect prompt injection.
Ingestion points: The prompt field within the JSON payload passed to the belt app run command in SKILL.md.
Boundary markers: There are no explicit delimiters or instructions to ignore embedded commands within the prompt text.
Capability inventory: The agent's capabilities are restricted to audio generation and media merging via the belt tool.
Sanitization: The instructions do not define methods for sanitizing or validating the input text before processing.

dialogue-audio