The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process natural language inputs which are passed directly to an external CLI tool. This creates a surface where malicious instructions could be embedded in the data.
Ingestion points: User-provided values for voice_script, voice_prompt, and video_prompt within belt app run commands.
Boundary markers: Absent. The skill does not use delimiters or specific instructions to separate user data from command parameters.
Capability inventory: The skill utilizes the belt CLI tool to execute remote inference tasks on the inference.sh platform.
Sanitization: No evidence of input validation or sanitization is present in the skill instructions.
[EXTERNAL_DOWNLOADS]: The skill references and encourages the installation of external components from the inference.sh service.
Recommends installing the belt CLI via npx skills add belt-sh/cli.
References installation documentation hosted at https://raw.githubusercontent.com/inference-sh/skills/refs/heads/main/cli-install.md.
[COMMAND_EXECUTION]: The skill instructions rely on executing shell commands via the belt CLI, including authentication (belt login) and running remote applications (belt app run).

ai-avatar-video