skill-creator

To structure and package AI skills, organize your project with a SKILL.md file, scripts, references, and assets, then use a validation tool to bundle it into a distributable .skill file. This ensures the skill is modular and easy for AI agents to use.

The best way to organize AI skill code and documentation is to separate resources into scripts for executable logic, references for static data, and assets for supporting files, all documented via a well-structured SKILL.md frontmatter and body.

You can initialize a new skill project with the correct file structure by running the `init_skill.py` script. This sets up the foundational directories and files needed for proper AI skill development and packaging.

Yes, you can validate your AI skill configuration before packaging it into a distributable file. The packaging process includes built-in validation checks to ensure your SKILL.md frontmatter and resource organization meet deployment standards.

To bundle your skill files into a .skill package for distribution, run the `package_skill.py` script. This tool automates the packaging process while applying validation rules to ensure the final archive is correctly formatted.

You should use scripts for executable code and active logic in your AI skill, whereas references should be used for static documentation or data that the agent reads. Proper resource organization keeps the skill modular and efficient.