command-development

Design Claude Code slash commands with Markdown, YAML frontmatter, and tool permissions.

1|Updated Mar 30, 2026
One-click install
npx skills add https://github.com/Leap0920/Clean-Portfolio --skill command-development-leap0920
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: command-development
Source: https://github.com/Leap0920/Clean-Portfolio/tree/main/%25USERPROFILE%25/.openclaude/plugins/marketplaces/claude-plugins-official/plugins/plugin-dev/skills/command-development
Command: npx skills add https://github.com/Leap0920/Clean-Portfolio --skill command-development-leap0920

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill helps you build reliable Claude Code slash commands by teaching the required Markdown structure, YAML frontmatter configuration, and the correct ways to pass inputs and resources into a command.

Core Features & Use Cases

  • Define command behavior with YAML frontmatter: Set discoverable descriptions, model selection, tool permissions, argument hints, and manual-only invocation.
  • Design flexible command interfaces: Capture dynamic user input with $ARGUMENTS and positional arguments like $1, $2, and validate expectations in the prompt.
  • Inject context safely and effectively: Load file contents into the command via @file references and gather live repository context using Bash execution (!... syntax).
  • Organize commands for teams and plugins: Use namespacing patterns for plugin commands and integrate command workflows with plugin components (agents, skills, hooks).
  • Use references and examples progressively: Start with core fundamentals and expand into advanced workflows, interactive commands (AskUserQuestion), testing strategies, and marketplace-ready documentation patterns.

Quick Start

Create a new command file and start with a YAML block that includes a description, then write the command instructions to guide Claude on exactly what to do when the slash command is invoked.

Frequently Asked Questions about command-development

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I create a Claude Code slash command with YAML frontmatter?

To create a Claude Code slash command, you write a Markdown file with a YAML frontmatter block that defines the command's description, model selection, and tool permissions, followed by Markdown instructions that guide Claude's behavior when the slash command is invoked.

How do I pass dynamic arguments to a custom slash command?

You can pass dynamic arguments to a custom slash command by using the $ARGUMENTS variable for full input capture or positional variables like $1 and $2, which Claude substitutes into the command prompt at runtime to process user-provided data.

Can I inject live repository context into a slash command using bash?

Yes, you can inject live repository context into a slash command by using the bash execution syntax (!`...`) in the Markdown prompt, which runs shell commands and feeds their output directly into Claude's context window for processing.

How do I load file contents into a Claude Code command prompt?

You load file contents into a Claude Code command prompt by using the @file reference syntax within the Markdown instructions, allowing the command to read and process specific file data dynamically during execution.

How do I restrict tool permissions for a custom slash command?

You restrict tool permissions for a custom slash command by defining the allowed-tools property in the YAML frontmatter, which gates bash execution and file references to ensure the command only accesses explicitly permitted tools.

Does this Skill support building multi-step workflows with interactive decisions?

Yes, this Skill supports building multi-step workflows that require interactive decisions by incorporating AskUserQuestion logic and progressive examples into the Markdown prompt, allowing Claude to pause and request user input during command execution.