Slash Command Factory

A comprehensive system for generating production-ready Claude Code slash commands through a simple question-based workflow.

Self-Evolving Skill: This skill improves through use. If instructions are wrong, parameters drifted, or a workaround was needed — fix this file immediately, don't defer. Only update for real, reproducible issues.

When to Use This Skill

• Creating new custom slash commands for Claude Code
• Generating command templates from presets
• Validating slash command YAML frontmatter syntax
• Organizing commands into proper folder structures
• Converting workflows into reusable slash commands

Overview

This skill helps you create custom slash commands for Claude Code by:

• Asking 5-7 straightforward questions about your command needs
• Generating complete command .md files with proper YAML frontmatter
• Providing 10 powerful preset commands for common use cases
• Validating command format and syntax
• Creating well-organized folder structures
• Offering installation guidance

Output: Complete slash commands ready to use in Claude Code

Command Structure Patterns

Three official patterns from Anthropic documentation:

Full templates and when-to-use guidance: Command Patterns

Naming Convention

All slash command files MUST follow kebab-case: [verb]-[noun].md

Rules: Lowercase only, 2-4 words, [a-z0-9-] characters, no underscores/camelCase

Full conversion algorithm and examples: Naming Convention

Bash Permission Rules

Critical: Blanket Bash permission is prohibited. Must use subcommand-level specificity.

# WRONG - too broad
allowed-tools: Bash(git:*)

# CORRECT - subcommand-level
allowed-tools: Bash(git add:*), Bash(git commit:*), Bash(git push:*)

# OK - simple commands without subcommand hierarchies
allowed-tools: Bash(cp:*), Bash(mkdir -p:*), Bash(date:*)

Full patterns and selection guide: Bash Permissions

Two Paths to Generate Commands

Path 1: Quick-Start Presets (30 seconds)

Choose from 10 powerful preset commands:

Full YAML configs and details: Preset Commands

Path 2: Custom Command (5-7 Questions)

Create a completely custom command by answering questions about:

1. Purpose - What should the command do?
2. Arguments - Auto-determined; all flags get mandatory short forms (-b|--branch)
3. Tools - Which Claude Code tools (Read, Write, Bash, Grep, Task, etc.)
4. Agents - Does it need to launch specialized agents?
5. Output - Analysis, files, action, or report?
6. Model - Default, Sonnet, Haiku, or Opus? (optional)
7. Features - Bash execution, file references, context gathering? (optional)

Full question scripts and argument conventions: Question Flow

Generation & Installation

After collecting answers, the skill:

1. Generates YAML frontmatter with proper allowed-tools
2. Generates command body with purpose-specific instructions
3. Creates folder structure under generated-commands/[command-name]/
4. Validates format (YAML, arguments, tools, organization)
5. Provides installation instructions

# Install to project
cp generated-commands/[command-name]/[command-name].md .claude/commands/

# Install globally
cp generated-commands/[command-name]/[command-name].md ~/.claude/commands/

Plugin invocation: /plugin-name:command-name [arguments]

Full process, folder structure, and plugin invocation rules: Generation Process

Validation

Every generated command is validated for:

• Valid YAML frontmatter (proper syntax, required fields)
• Correct argument format ($ARGUMENTS, not $1 $2 $3)
• Short forms for all flags (mandatory 1-2 letter shortcuts)
• Bash subcommand-level specificity (no blanket Bash)
• Clean folder organization

If validation fails, you get specific fix instructions.

Full validation checklist, best practices, and troubleshooting: Validation Reference

Quick Reference

Usage

@slash-command-factory
Use the /research-business preset

@slash-command-factory
Create a custom command for analyzing customer feedback

Key Rules

Ecosystem Integration

Works with: factory-guide, skills-guide, prompts-guide, agents-guide

More examples and integration details: Usage Examples

Post-Execution Reflection

After this skill completes, check before closing:

1. Did the command succeed? — If not, fix the instruction or error table that caused the failure.
2. Did parameters or output change? — If the underlying tool's interface drifted, update Usage examples and Parameters table to match.
3. Was a workaround needed? — If you had to improvise (different flags, extra steps), update this SKILL.md so the next invocation doesn't need the same workaround.

Only update if the issue is real and reproducible — not speculative.