The Agent Skills Directory

[SAFE]: The skill consists entirely of natural language instructions designed to guide the AI's writing style. It does not include any scripts, shell commands, or executable logic.
[SAFE]: The skill does not perform any network requests, file system modifications, or credential handling. All operations are confined to the AI's internal text processing context.
[SAFE]: The frontmatter configuration in SKILL.md correctly limits the agent's behavior by disabling further model invocations while processing the rewrite, reducing the risk of unintended behavior.
[SAFE]: The agents/openai.yaml file defines standard interface parameters and policies that align with the skill's stated purpose without introducing additional capabilities or risks.

qstack-unslop