template-skill

To scaffold a new AI agent skill, you create a SKILL.md file with a standardized YAML frontmatter containing name and description fields. This provides consistent discovery metadata for automatic onboarding.

YAML frontmatter in skill scaffolding exposes the name and description metadata required for automatic discovery by AI agents. It ensures every new skill is consistently indexed and identifiable across the ecosystem.

You create a minimal execution body by filling the SKILL.md template below the YAML frontmatter with actionable operational instructions. These instructions are loaded on activation when the agent invokes the skill.

You do not need to include scripts, references, or assets directories when scaffolding skills. These are optional directories allowed by the template for future extensions once the base skill is operational.

You can use this scaffolding template for automations, tooling, and agent behaviors. It simplifies the creation of new skills by providing a ready-to-edit structure that reduces boilerplate time for developers.

The best way to reduce boilerplate when building AI agent skills is using a standardized scaffolding process with a ready-to-edit SKILL.md frontmatter. This ensures consistent discovery metadata and a minimal body for rapid onboarding.