skill-creator

Create a Claude skill by using the standardized template structure: define YAML frontmatter with name and description, write a concise SKILL.md for orchestration, and bundle scripts, references, and assets. Run `python scripts/init_skill.py skill-name --path skills/public` to scaffold a new skill with the required structure.

Structure skills modularly using SKILL.md as the orchestration body, organize metadata in YAML frontmatter, and bundle related resources—scripts, references, and assets—together. This approach enables progressive disclosure loading and ensures skills are reusable and maintainable across different workflows.

Yes. Skill templates accelerate development by providing predefined SKILL.md structures, versioning conventions, ownership metadata, and quality checks. Templates support workflow-based, task-based, and reference-style skills, eliminating the need to build structure from scratch.

Include YAML frontmatter (name, description), a SKILL.md orchestration file, optional scripts for automation, reference materials, and assets. Bundle these together with versioning and ownership information to create a complete, quality-checked skill package.

Organize scripts and assets as bundled resources within the skill's directory structure. Scripts enable reliable automation and tool integration, while assets and references provide context. Progressive disclosure loading allows users to access only what they need for their specific workflow.

Define skill metadata in YAML frontmatter including name, description, versioning, and ownership. Include functional and technical requirements specifying what the skill does, its dependencies, supported workflows, and integrated quality checks for consistency and reliability.