skill-creator

To create modular skill units for agents, run the init_skill.py script to initialize a new directory, then customize the mandatory SKILL.md frontmatter with name and description, and add optional scripts, references, or assets folders as needed.

A valid SKILL.md frontmatter requires name and description fields to pass spec validation. Optional resource directories like scripts, references, and assets can be bundled alongside the SKILL.md file to support complex agent workflows.

You validate and package automation skills using the package_skill.py script, which enforces the Skill Unit spec by checking for mandatory SKILL.md frontmatter and bundling optional resource directories into a deployable format.

Yes, you can bundle external scripts and references into a skill unit by placing them in the optional scripts/ and references/ directories, allowing agents to access supporting tooling and documentation for complex tasks.

No external dependencies are required to maintain skill templates. The toolset provides self-contained Python scripts for initialization and packaging, handling validation and modular design natively without external libraries.

The best way to structure modular automation tooling is using self-contained Skill Units with a mandatory SKILL.md file, leveraging the provided init_skill.py script to scaffold the directory and enforce the required frontmatter validation.