Can I pass dynamic arguments to my custom commands?

Yes, you can capture user input using $ARGUMENTS for the entire input string or positional variables like $1, $2, and $3 for specific arguments.

How do I create a custom slash command in Claude Code?

Custom commands are created by adding Markdown files (.md) to the `.claude/commands/` directory for project-specific commands or `~/.claude/commands/` for personal, global commands.

What is the purpose of YAML frontmatter in a command file?

YAML frontmatter allows you to configure command metadata such as the description shown in help menus, which AI model to use, and which tools (like Bash or Read) the command is allowed to access.

How can I include project context automatically in a command?

You can use the '@' syntax to reference specific files or use inline bash execution (wrapped in backticks and starting with '!') to run commands like 'git diff' to gather context before Claude processes the prompt.

Claude Code Command Development

Name: Claude Code Command Development
Author: liminal-ai

byliminal-ai

0•

开发者工具

Guides the creation and optimization of custom slash commands to automate workflows and standardize AI interactions.

This skill provides a comprehensive framework for developing custom slash commands within Claude Code, allowing developers to extend the AI's capabilities through reusable, structured prompts. It covers everything from basic Markdown command files and YAML frontmatter configuration to advanced features like dynamic positional arguments, inline bash execution for context gathering, and hierarchical command organization. By focusing on instruction-based prompts designed for agent consumption, this skill helps teams build consistent, efficient, and sharable development workflows.

主要功能

01Integrate real-time context via inline bash command execution

02Organize commands with flat or namespaced directory structures

03Master YAML frontmatter fields like description, allowed-tools, and model selection

040 GitHub stars

05Apply best practices for agent-centric prompt engineering and file referencing

06Implement dynamic argument handling using $ARGUMENTS and positional variables

使用场景

01Developing reusable documentation generators that reference specific source files

02Building project-specific shortcuts for testing, building, and deployment

03Creating standardized code review and security auditing workflows

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add liminal-ai/code-steward command-development

For use in Claude.ai and ChatGPT

主要功能

01Integrate real-time context via inline bash command execution

02Organize commands with flat or namespaced directory structures

03Master YAML frontmatter fields like description, allowed-tools, and model selection

040 GitHub stars

05Apply best practices for agent-centric prompt engineering and file referencing

06Implement dynamic argument handling using $ARGUMENTS and positional variables

使用场景

01Developing reusable documentation generators that reference specific source files

02Building project-specific shortcuts for testing, building, and deployment

03Creating standardized code review and security auditing workflows

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add liminal-ai/code-steward command-development

For use in Claude.ai and ChatGPT