The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill points to an external installation document located at https://raw.githubusercontent.com/inference-sh/skills/refs/heads/main/cli-install.md. This resource originates from an external repository that is not included in the trusted provider lists.
[REMOTE_CODE_EXECUTION]: The skill recommends adding further capabilities using npx skills add inference-sh/skills@... commands. These commands download and install external code packages into the agent's operating environment from a non-standard source.
[INDIRECT_PROMPT_INJECTION]: The skill presents a surface for indirect prompt injection attacks:
Ingestion points: User-controlled data enters the agent context through the text parameter in the belt app run JSON payloads described in the examples.
Boundary markers: Absent. The instructions do not specify delimiters or provide guidance to the agent to isolate the text input from its instruction set.
Capability inventory: The skill possesses the capability to execute the belt CLI tool via a scoped Bash(belt *) permission.
Sanitization: Absent. There is no evidence of input validation, escaping, or filtering before the user-supplied text is passed to the shell command.

elevenlabs-tts