The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill directs users to install the belt-sh/cli and references an installation guide hosted on the author's official GitHub repository. These resources are part of the vendor's intended platform functionality.
[COMMAND_EXECUTION]: The skill provides examples for using the belt CLI to execute specific research and image generation tasks. The commands are well-defined and align with the persona creation use case without attempting to execute arbitrary or hidden code.
[PROMPT_INJECTION]: The skill creates an attack surface for indirect prompt injection because it processes content retrieved from external search engines (Tavily, Exa).
Ingestion points: Search results retrieved via belt app run from Tavily and Exa are returned to the agent context (SKILL.md).
Boundary markers: No explicit delimiters or instructions to ignore embedded commands are present in the provided command examples.
Capability inventory: The skill is restricted to the Bash(belt *) toolset.
Sanitization: Input sanitization is not explicitly handled within the skill's instructions, relying instead on the agent's internal safety filters.

customer-persona