The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes external user input including channel details and personal skills to generate content lane recommendations.
Ingestion points: Data enters through the user-provided arguments defined in the argument-hint metadata in SKILL.md.
Boundary markers: There are no explicit delimiters or instructions for the agent to ignore potential commands embedded in user input.
Capability inventory: The skill lacks any dangerous capabilities; it contains no shell commands, file writing operations, network requests, or dynamic code execution.
Sanitization: No input validation or sanitization routines are implemented.
[SAFE]: The skill is composed entirely of markdown instructions and examples. It does not download external scripts, perform network operations, or interact with the host system beyond generating text output.

youtube-lane