Slash Command Builder

Slash Command Builder - Claude Code Command Expert

Use this skill when creating, improving, or troubleshooting Claude Code slash commands. Provides expert guidance on command structure, syntax, frontmatter configuration, and best practices.

When to Use This Skill

Activate this skill when:

• User asks to create a new slash command
• User wants to improve an existing command
• User needs help with command arguments or frontmatter
• User is troubleshooting command invocation issues
• User wants to understand slash command capabilities
• User asks about the difference between commands and skills

Quick Reference

Command File Structure

---
description: Brief description shown in autocomplete
argument-hint: [arg1] [arg2] <optional-arg>
allowed-tools: Bash(git *), Read, Write
model: claude-sonnet-4-6
disable-model-invocation: false
---

Your command prompt here with $ARGUMENTS or $1, $2, etc.

Use !`command` for bash execution
Use @file.txt for file references

File Locations

Project commands (shared with team):

.claude/commands/my-command.md

Personal commands (individual use):

~/.claude/commands/my-command.md

Organized commands (namespaced):

.claude/commands/frontend/component.md  → /component (project:frontend)

Creating Effective Slash Commands

Step 1: Identify the Use Case

Good candidates for slash commands:

• Frequently repeated prompts
• Simple, single-purpose tasks
• Quick code reviews or analyses
• Common git workflows
• Standard documentation tasks

NOT good for slash commands (use Skills instead):

• Complex multi-step workflows
• Context-aware behavior
• Team standardization needs
• Workflows requiring multiple files

Step 2: Choose Command Name

Best practices:

• Use lowercase with hyphens: /review-pr, /optimize-code
• Make it memorable and intuitive
• Avoid conflicts with built-in commands
• Keep it short (2-3 words max)

Examples:

• ✅ /review-pr - Clear, concise
• ✅ /fix-lint - Action-oriented
• ❌ /review-pull-request-thoroughly - Too verbose
• ❌ /rpr - Too cryptic

Step 3: Design the Prompt

Simple command (no arguments):

---
description: Analyze code for performance bottlenecks
---

Analyze the current file for performance issues:
1. Identify O(n²) or worse algorithms
2. Find unnecessary re-renders or computations
3. Check for memory leaks
4. Suggest optimizations with code examples

Command with all arguments:

---
description: Generate component boilerplate
argument-hint: <component-name> <type>
---

Create a $ARGUMENTS component following our style guide:
- Use TypeScript with strict types
- Include prop interfaces
- Add JSDoc comments
- Export as default

Command with positional arguments:

---
description: Review pull request
argument-hint: [pr-number] [reviewer]
---

Review PR #$1 and assign to @$2:
1. Check code quality and style
2. Verify tests are included
3. Look for security issues
4. Suggest improvements
5. Add comments in GitHub

Step 4: Add Frontmatter Configuration

See FRONTMATTER.md in this skill directory for complete frontmatter options.

Minimal frontmatter:

---
description: What this command does
---

Full-featured frontmatter:

---
description: Complete command with all options
argument-hint: [required] <optional>
allowed-tools: Bash(git *), Read(**/*.ts), Write
model: claude-sonnet-4-6
disable-model-invocation: false
---

Step 5: Test the Command

1. Save the file to .claude/commands/
2. Invoke with /command-name
3. Check argument substitution works
4. Verify tool permissions if using allowed-tools
5. Test edge cases (missing args, wrong types)

Advanced Features

Bash Execution

Execute shell commands inline with ! prefix:

---
description: Show git status
allowed-tools: Bash(git status:*)
---

Current repository status:

!`git status`

Recent commits:

!`git log --oneline -5`

File References

Include file contents with @ prefix:

---
description: Review specific file
argument-hint: <file-path>
---

Review this file for code quality:

@$1

Focus on:
- Type safety
- Error handling
- Performance
- Maintainability

Tool Permissions

Restrict which tools Claude can use:

---
description: Safe git status check
allowed-tools: Bash(git status:*), Bash(git diff:*)
---

Show current changes:

!`git status`
!`git diff --stat`

Tool permission syntax:

• Bash(command:*) - Allow specific command with any args
• Read(path/to/*.ts) - Allow reading TypeScript files in path
• Write - Allow writing any file
• Glob, Grep, Edit - Other available tools

Model Selection

Override default model for specific commands:

---
description: Quick syntax fix
model: claude-haiku-4-5-20251001
---

Fix syntax errors in the current file quickly.

When to use different models:

• claude-haiku-4-5-20251001 - Fast, simple tasks
• claude-sonnet-4-6 - General purpose (default)
• claude-opus-4-8 - Complex reasoning

Disable Auto-Invocation

Prevent Claude from calling command automatically:

---
description: Destructive operation
disable-model-invocation: true
---

!`rm -rf node_modules`
!`npm install`

Common Patterns

Code Review Command

---
description: Review code changes
argument-hint: [file-or-pr]
allowed-tools: Bash(git *), Read, Grep
---

Review $ARGUMENTS for:

1. **Code Quality**
   - Clean, readable code
   - Proper naming conventions
   - DRY principle

2. **Security**
   - Input validation
   - SQL injection risks
   - XSS vulnerabilities

3. **Performance**
   - Inefficient algorithms
   - Unnecessary computations
   - Memory leaks

4. **Tests**
   - Unit test coverage
   - Edge cases handled
   - Integration tests

Provide specific file:line references for all issues.

Git Workflow Command

---
description: Create feature branch
argument-hint: <feature-name>
allowed-tools: Bash(git *)
---

Create and switch to feature branch:

!`git checkout -b feature/$1`
!`git push -u origin feature/$1`

Branch feature/$1 created and pushed to origin.

Documentation Generator

---
description: Generate API docs
argument-hint: <file-path>
allowed-tools: Read
---

Generate comprehensive API documentation for:

@$1

Include:
- Function signatures with types
- Parameter descriptions
- Return value documentation
- Usage examples
- Error cases

Test Generator

---
description: Generate test cases
argument-hint: <file-to-test>
allowed-tools: Read, Write
---

Generate test cases for:

@$1

Create tests covering:
- Happy path scenarios
- Edge cases
- Error conditions
- Boundary values

Use the existing test framework style.

Troubleshooting

Command Not Found

Problem: /my-command doesn't autocomplete

Solutions:

1. Check file is in .claude/commands/ or ~/.claude/commands/
2. Verify filename matches command (.claude/commands/my-command.md)
3. Restart Claude Code to reload commands
4. Check for syntax errors in frontmatter

Arguments Not Substituting

Problem: $1 appears literally instead of being replaced

Solutions:

1. Ensure arguments are passed: /command arg1 arg2
2. Check you're using $1, $2 (not ${1})
3. For all args, use $ARGUMENTS instead
4. Verify argument-hint frontmatter is correct

Tool Permission Denied

Problem: Command can't execute bash or read files

Solutions:

1. Add allowed-tools frontmatter
2. Use specific tool patterns: Bash(git *)
3. Check tool name capitalization (e.g., Bash, not bash)
4. For file operations, use glob patterns: Read(**/*.ts)