The Agent Skills Directory

[PROMPT_INJECTION]: The skill exhibits a surface for indirect prompt injection by interpolating untrusted data into command arguments for voice synthesis.
Ingestion points: User-provided text is passed directly into the prompt and text fields of JSON payloads used with the belt app run command in SKILL.md.
Boundary markers: The skill does not use delimiters or provide instructions to the agent to ignore potential commands embedded within the text to be spoken.
Capability inventory: The skill utilizes the belt CLI tool to perform audio generation, media merging, and video animation tasks.
Sanitization: No sanitization, escaping, or validation of the input text is performed before it is passed to the execution environment.
[EXTERNAL_DOWNLOADS]: The skill references installation instructions and configuration files from the inference.sh official domain and public GitHub repository.
Evidence: References https://raw.githubusercontent.com/inference-sh/skills/refs/heads/main/cli-install.md and suggests installing the belt-sh/cli package via npx.
[COMMAND_EXECUTION]: The skill relies on the execution of the belt CLI tool to perform its primary functions, including application execution (belt app run) and session management (belt login).

ai-voice-cloning