skill-creator

To create reusable AI agent skills with standardized governance, generate a new skill directory containing a SKILL.md file with required YAML frontmatter, operational instructions, and organizational naming conventions to ensure consistent AI loading.

The best way to structure workflow templates for AI agents is to separate static schemas into an assets directory and link local documentation into a references directory, encoding decision rules directly within the operational instructions of SKILL.md.

You should create a new skill when repeated patterns need standardized guidance, project conventions require strict documentation, and workflow decision rules must be encoded for AI loading, rather than relying on simple reference documents.

The required YAML frontmatter fields for an AI-ready SKILL.md file are name, description, license, and metadata, which enforce a consistent Skill structure and satisfy discoverability requirements for agent loading.

Document project conventions for AI agent workflow execution by drafting a governance-first SKILL.md entry that encodes workflow decision rules, applies standardized naming, and integrates assets or references for repeatable patterns.

Yes, you can use local documentation links inside a skill directory structure by placing them in the references directory, reserving the assets directory for templates and schemas needed during workflow execution.