Claude Skills Cli

claude-skills-cli

CLI for creating Claude Agent Skills with progressive disclosure validation. Built for Claude Code to use when humans ask it to create skills.

For Humans Using Claude Code

Tell Claude Code to create skills with specific package manager commands.

Using npm:

"Use npx -y claude-skills-cli to create a skill for my API client code"

Using pnpm:

"Use pnpx claude-skills-cli to create a skill for my API client code"

This CLI enforces progressive disclosure so Claude doesn't write 500-line documentation files.

Progressive Disclosure

Skills load in 3 levels to optimize token usage:

Level	Content	When Loaded	Size Limit
1	SKILL.md metadata (YAML)	Always in context	<200 chars
2	SKILL.md body (Markdown)	When skill triggers	~50 lines
3	references/, scripts/, assets/	As needed by Claude	Unlimited

This CLI enforces these constraints through validation, keeping skills lean and scannable.

Commands

init

pnpx claude-skills-cli init --name my-skill --description "Brief description with trigger keywords"
pnpx claude-skills-cli init --name my-skill --description "..." --with-examples

Creates minimal skill scaffolding by default (SKILL.md, references/). Use `--with-examples` for example files.

validate

pnpx claude-skills-cli validate .claude/skills/my-skill
pnpx claude-skills-cli validate .claude/skills/my-skill --lenient
pnpx claude-skills-cli validate .claude/skills/my-skill --loose
pnpx claude-skills-cli validate .claude/skills/my-skill --strict
pnpx claude-skills-cli validate .claude/skills/my-skill --format json

**Validation Modes:**

Mode	Max Lines	Use Case
(default)	50	Strict best practices
--lenient	150