The Agent Skills Directory

[SAFE]: The skill consists entirely of markdown documentation providing stylistic guidelines for writing and brand voice consistency. It does not perform any functional operations on the system or network.
[NO_CODE]: No scripts, binaries, or executable commands are present. The configuration in SKILL.md explicitly restricts the skill to reference-only mode by disabling model invocation and allowing no tools.
[PROMPT_INJECTION]: The instructions are strictly limited to content creation rules, such as avoiding filler words and exclamation points. There are no patterns suggesting attempts to bypass safety filters or override agent behavior.

kernel-voice