Guide the creation of AI Skills with SKILL.md structure and validation.
npx skills add https://github.com/johnmog/dotfiles --skill skill-creator-johnmog
Please help me install this Agent Skill. Skill: skill-creator Source: https://github.com/johnmog/dotfiles/tree/main/.copilot/skills/skill-creator Command: npx skills add https://github.com/johnmog/dotfiles --skill skill-creator-johnmog
This Skill provides comprehensive guidance and best practices for creating new AI Skills or updating existing ones, ensuring they are effective, efficient, and well-structured.
SKILL.md file and optional scripts/, references/, and assets/ directories.Use the skill-creator to learn how to build a new AI skill.
High-intent search queries and answers about installing and using this skill.
Progressive disclosure in AI skill creation is a design principle that structures skill content to reveal information sequentially. You should use it when building modular extensions to keep the AI agent focused and prevent context overload.
To structure an AI skill, create a required SKILL.md file for core logic and use optional scripts, references, and assets directories for bundled resources. This modular structure ensures the AI agent can efficiently access necessary components.
Best practices for writing SKILL.md content include maintaining conciseness, setting appropriate degrees of freedom for the AI agent, and validating the structure. Proper frontmatter ensures the skill is discoverable and usable by the agent.
Yes, you can use bundled scripts and references to extend AI agent capabilities. These optional directories allow you to package external assets and logic alongside the SKILL.md file, creating robust modular extensions without cluttering the core prompt.
To validate a new AI skill, follow the creation process guide from understanding the skill to iteration. Validation ensures the SKILL.md content, frontmatter, and bundled resources function correctly and are fully discoverable by the AI agent.
tyk-labCreate an isolated git worktree for feature development with safety checks.
mistakenotExecute multi-agent AI programs defined in prose files.
wunkiCreates and maintains AI skills by structuring files and writing SKILL.md content.
mpazarynaCreate and validate Goose recipe YAML/JSON configurations for AI agent workflows.