The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill requires the installation of external tools and packages from sources not identified as trusted.
Instructs the user to run npx skills add belt-sh/cli to install the command-line interface.
Recommends adding multiple external skills from the inference-sh organization.
[COMMAND_EXECUTION]: The skill's primary functionality relies on executing shell commands through the belt CLI.
Example: belt app run falai/flux-dev-lora --input '...' is used to invoke remote AI models.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection through the interpolation of variables into shell command arguments.
Ingestion points: Shell variables like $STYLE defined in SKILL.md are directly interpolated into model prompts.
Boundary markers: No delimiters or instructions to ignore embedded commands are present in the interpolation templates.
Capability inventory: The skill has the capability to execute remote applications and process images via belt app run.
Sanitization: The skill does not perform any escaping or validation of the content within the variables before execution.

storyboard-creation