Can I execute terminal commands within a slash command?

Yes, you can use the '!' prefix to execute bash commands before the prompt runs, allowing you to gather context like git status or test results.

What is the benefit of the allowed-tools field?

The allowed-tools field provides security and scope control by restricting which tools a command can access, preventing accidental file modifications or unauthorized executions.

Does this skill provide official Claude Code syntax?

It builds upon the official specification, offering both the standard syntax and opinionated best practices to ensure high-quality command implementations.

When should I use a command instead of an agent?

Use a command when you want explicit user control over a deterministic task. Use an agent when you need isolated context, parallelization, or autonomous delegation.

How do I handle multiple arguments in a command?

Use $1, $2, etc., for positional arguments that require distinct semantic logic, or $ARGUMENTS to pass the entire input string to a prompt or sub-agent.

Slash Command Design

Name: Slash Command Design
Author: racurry

byracurry

0•

Developer Tools

Provides expert guidance and best practices for creating, structuring, and optimizing Claude Code slash commands.

This skill serves as a comprehensive interpretative guide for developers building specialized commands within Claude Code. It bridges the gap between official documentation and practical implementation by offering a structured decision framework for choosing between commands, agents, or skills. By following its opinionated guidance on argument patterns, tool restrictions, and delegation to sub-agents, developers can create robust, efficient, and highly discoverable command interfaces that enhance their local development environment.

Key Features

01Instructions for advanced features like bash execution (!) and file references (@)

020 GitHub stars

03Decision framework for component selection (Command vs. Agent vs. Skill)

04Checklists for validating command quality and avoiding common anti-patterns

05Guidance on positional ($1) vs. full-string ($ARGUMENTS) patterns

06Best practices for frontmatter fields and security-conscious tool restrictions

Use Cases

01Implementing secure, restricted commands for sensitive repository operations

02Designing custom slash commands for project-specific automation and workflows

03Reviewing and refactoring existing commands to improve efficiency and isolation

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add racurry/neat-little-package slash-command-design

For use in Claude.ai and ChatGPT

Download Skill