The Agent Skills Directory

[SAFE]: The skill does not contain any malicious patterns, obfuscation, or unauthorized data access attempts. It serves as a legitimate development tool for building other skills.
[COMMAND_EXECUTION]: The skill documentation describes a structure for including executable scripts (.sh) within a skill's directory and includes the Task tool in its allowed-tools. This is documented as a feature for deterministic workflows and is consistent with the skill's purpose as a development utility.
[PROMPT_INJECTION]: No attempts to bypass safety filters, override system instructions, or extract system prompts were found. The workflow focus is on structural design rather than behavioral manipulation.
[REMOTE_CODE_EXECUTION]: There are no patterns involving the download or execution of remote scripts from external URLs. All referenced paths are local to the skill's directory.

create-skill