The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill references and fetches content from external sources, specifically the inference-sh GitHub repository for installation instructions.
Evidence: Reference to https://raw.githubusercontent.com/inference-sh/skills/refs/heads/main/cli-install.md and instructions to install the belt-sh/cli skill via npx.
[COMMAND_EXECUTION]: The skill documents the use of the belt CLI tool for authenticating and running dubbing applications.
Evidence: Usage of belt login and belt app run elevenlabs/dubbing commands throughout the SKILL.md examples.
[INDIRECT_PROMPT_INJECTION]: The skill ingests external data in the form of audio and video URLs which are passed to the ElevenLabs dubbing engine. This is a standard input surface for the intended media translation purpose.
Ingestion points: audio and video_url fields in the JSON input for the belt commands (SKILL.md).
Boundary markers: The input is structured as JSON, which provides clear boundaries for data values.
Capability inventory: The skill uses shell commands via the Bash(belt *) tool to execute remote applications.
Sanitization: Not explicitly implemented in the skill documentation, relying on the underlying belt CLI and ElevenLabs engine safety filters.

elevenlabs-dubbing