skill-creator

To create a Claude skill with standardized YAML frontmatter, use a creation framework that enforces name and description metadata, scaffolds scripts and references, and packages resources into a ready-to-use unit. This skill guides that initialization workflow.

The best way to package modular Claude skills is to enforce YAML frontmatter validation with name and description fields, organize optional scripts and references, and apply progressive disclosure to keep the SKILL.md lean while linking to detailed assets.

Yes, the skill creation workflow supports adding custom scripts, references, and assets. It provides templates and example resources to jumpstart creation, allowing you to include optional components while maintaining standardized metadata.

The toxicity check guard works during skill creation by enforcing a validation step that screens the skill's content. It ensures the generated skill units meet safety standards before the initialization and packaging workflows finalize the output.

Yes, you can use PyYAML to validate frontmatter in generated Claude skills. The skill creation framework relies on PyYAML as a dependency to parse and enforce the required name and description fields in the YAML metadata.

A Claude skill needs a lean SKILL.md with progressive disclosure to keep the core file readable while linking to detailed references. This approach ensures consistent discovery without overwhelming the user with embedded scripts and asset details.