The Agent Skills Directory

[SAFE]: The skill serves as a legitimate framework for skill development, consisting of markdown documentation and local Python utility scripts.- [COMMAND_EXECUTION]: The skill includes local scripts scripts/init_skill.py and scripts/package_skill.py to automate file system tasks:
init_skill.py: Creates directory structures and template files based on user-provided names.
package_skill.py: Bundles skill directories into ZIP archives using the standard zipfile library.
These scripts operate strictly on the local file system and do not perform network requests.- [INDIRECT_PROMPT_INJECTION]: The skill functions as a template generator that incorporates user-provided metadata (name, description) into generated files.
Ingestion points: User-supplied arguments to the init_skill.py command-line interface.
Boundary markers: None explicitly defined in the templates, though content is separated into distinct files.
Capability inventory: File system write access and permission modification (chmod 755) for generated example scripts in scripts/init_skill.py.
Sanitization: scripts/quick_validate.py provides validation logic that enforces hyphen-case naming conventions and restricts the use of angle brackets in descriptions to mitigate basic injection risks.

skill-creator