Introducing Commands: Custom Slash Commands for Claude Code

We're excited to announce Commands—a new feature for creating reusable slash commands that work in Claude Code.

What Are Commands?

Commands are custom prompts that you can trigger using slash notation. Instead of typing the same instructions repeatedly, create a command once and invoke it with /command-name.

Each command consists of:

• Name — The slash command identifier (e.g., review, security-check)
• Description — What the command does
• Content — The prompt sent to your AI assistant
• Arguments — Optional placeholders for dynamic input

Think of commands as shortcuts for your most-used prompts.

Why Commands Matter

Consistency Across Your Team

When everyone uses the same command, everyone gets the same quality output. No more variations in how different developers ask for code reviews or security checks.

Save Time on Repetitive Tasks

How often do you type "review this code for security vulnerabilities, check for..."? With commands, that becomes /security-check. One trigger, complete instructions.

Standardize Best Practices

Encode your team's expertise into commands. Your senior developer's thorough review process becomes a command everyone can use.

How to Create a Command

1. Navigate to Commands

Go to your workspace and click the Commands tab. Click New Command.

2. Define the Basics

3. Add Arguments (Optional)

Commands support up to 3 arguments. Arguments are placeholders in your content that users fill in when invoking the command.

Example command content:

Review this {{language}} code for:
- Security vulnerabilities
- Performance issues
- Best practices for {{framework}}

Focus especially on {{focus_area}}.

Arguments defined:

• language — Programming language
• framework — Framework being used
• focus_area — Specific area to emphasize

4. Save and Use

Click Create and your command is ready. Enable it to make it available in Claude Code.

Using Commands

Once enabled, commands appear in Claude Code.

In Claude Code

Claude Code sees your enabled commands through SuperClawd. Simply mention the command or use slash notation to invoke it.

With Arguments

When you invoke a command with arguments, you'll be prompted to provide values:

/security-check
  language: TypeScript
  framework: React
  focus_area: authentication flows

The AI receives the complete prompt with your values inserted.

Example Commands

Code Review

Name: review
Description: Comprehensive code review

Content:
Review this code for:
1. Logic errors and edge cases
2. Security vulnerabilities
3. Performance optimizations
4. Code style and readability
5. Test coverage gaps

Provide specific, actionable feedback with code examples.

Security Audit

Name: security-check
Description: Security-focused code analysis

Content:
Analyze this code for security issues:

- Input validation and sanitization
- Authentication and authorization
- SQL injection and XSS vulnerabilities
- Sensitive data exposure
- Dependency vulnerabilities

Rate each finding by severity (Critical/High/Medium/Low).

Explain Code

Name: explain
Description: Detailed code explanation

Content:
Explain this code in detail:

1. What it does at a high level
2. How it works step by step
3. Any patterns or techniques used
4. Potential improvements or concerns

Assume I'm a {{experience_level}} developer.

Generate Tests

Name: test
Description: Generate unit tests

Content:
Generate comprehensive unit tests for this code:

- Cover happy path scenarios
- Include edge cases
- Test error handling
- Use {{test_framework}} syntax

Include setup, assertions, and cleanup as needed.

Export and Import

Export as JSON

Export any command as a JSON file for backup or sharing. Click the menu on a command card and select Export as JSON.

The JSON format:

{
  "name": "security-check",
  "description": "Security-focused code analysis",
  "content": "Analyze this code for security issues...",
  "arguments": [
    {
      "name": "severity_threshold",
      "description": "Minimum severity to report"
    }
  ]
}

Import Commands

Click Import next to New Command, then drag and drop your JSON file. The command is validated and added to your workspace.

If a command with the same name exists, a suffix is added automatically (e.g., review-2).

Community Commands

Browse and share commands with the developer community.

Discover Commands

Visit github.com/superclawd-ai/superclawd to find commands created by other developers. Download any JSON file and import it into your workspace.

Contribute Your Own

Have a command that works well? Share it!

1. Fork the repository
2. Add your command JSON to commands/examples/
3. Update the README catalog
4. Submit a pull request

All contributions are validated automatically to ensure they meet quality standards.

Commands vs Skills

Both Commands and Skills help your AI assistant, but they serve different purposes:

Use Commands for: Code reviews, generating tests, explaining code, refactoring tasks.

Use Skills for: Coding standards, style guides, project conventions.

Limits

• Name: Up to 30 characters, lowercase with hyphens
• Description: Up to 200 characters
• Content: Up to 5,000 characters
• Arguments: Maximum 3 per command
• Workspace: 1 command on free, up to 50 on paid

Get Started

Commands are available now in all SuperClawd workspaces.

1. Go to app.superclawd.com
2. Navigate to the Commands tab
3. Click New Command or Import
4. Start building your command library

Have feedback or ideas for Commands? Join the GitHub Discussions or reach out through our support page.

Pro tip: Check out the Community Commands for ready-to-import examples!