The Agent Skills Directory

[SAFE]: The skill facilitates the use of the ElevenLabs dubbing API via the vendor's own command-line interface tool (belt). All shell commands are structured as standard tool invocations.
[EXTERNAL_DOWNLOADS]: The skill references installation instructions and assets hosted on vendor-controlled infrastructure, including inference.sh domains and the official inference-sh GitHub organization. These resources are used for legitimate setup and documentation purposes.
[INDIRECT_PROMPT_INJECTION]: The skill ingests external data in the form of audio and video URLs provided by the user. This is a standard functional requirement for media processing, and the skill does not contain instructions that would allow these inputs to influence the agent's logic or override safety protocols.
[COMMAND_EXECUTION]: The skill uses the Bash tool to execute belt CLI commands. The allowed-tools configuration specifically restricts execution to the belt command, which is a security best practice that limits the attack surface.

elevenlabs-dubbing