Structure SKILL.md files with progressive disclosure and discovery principles.
npx skills add https://github.com/lidessen/moniro --skill authoring-skills-lidessen
Please help me install this Agent Skill. Skill: authoring-skills Source: https://github.com/lidessen/moniro/tree/main/skills/authoring-skills Command: npx skills add https://github.com/lidessen/moniro --skill authoring-skills-lidessen
This Skill helps authors design efficient, discoverable Agent Skills by enforcing progressive disclosure, concise explanations, and clear discovery signals.
To start, draft a concise SKILL.md frontmatter with name and description, place core workflow in SKILL.md, and move details to one or more reference files (one level deep). Keep SKILL.md under 500 lines and ensure references are clearly linked from SKILL.md.
High-intent search queries and answers about installing and using this skill.
Structure an Agent Skill by placing core workflows in SKILL.md and moving details to reference files one level deep. Keep SKILL.md under 500 lines and link references clearly to maximize token efficiency.
Progressive disclosure in skill authoring separates concise high-level logic from detailed references. Apply it by drafting a brief frontmatter and core workflow in SKILL.md, then moving granular details to linked reference files.
Best practices for organizing multi-file skills involve using clear naming conventions, keeping SKILL.md concise with frontmatter, and storing details in one-level-deep reference files linked directly from the main SKILL.md.
Yes, you can test Agent Skills. The skill authoring guide specifies testing strategies to validate discoverability, clear discovery signals, and predictable Claude behavior within your structured SKILL.md and references.
Yes, SKILL.md files should be kept under 500 lines. Exceeding this limit reduces token efficiency and hinders predictable Claude behavior, so you must move excess details into linked reference files.
DYAI2025Describes best practices for structuring SKILL.md with YAML frontmatter and XML body.
jon23dGuide AI skill creation, testing, and refinement with a TDD-inspired methodology.
luanCreate, edit, and verify AI skills using a TDD red-green-refactor cycle.
danielmiesslerOptimize AI agent prompts with structured Markdown and context engineering principles.
