Command Creator
Design and ship reusable Claude Code slash commands so repetitive dev workflows run as one `/command-name` invocation.
Overview
Command Creator is an agent skill most often used in Build (also Ship, Operate) that guides you through designing, implementing, and testing optimized Claude Code slash commands for reusable workflow automation.
Install
npx skills add https://github.com/softaworks/agent-toolkit --skill command-creatorWhat is this skill?
- End-to-end flow: design → implement → test agent-executable slash commands
- Covers command structure, location strategy (global vs project), and bundled resource patterns
- Documents command pattern catalog for reliable autonomous execution
- Best practices for clear instructions and optimal tool usage in commands
- Common use cases: repetitive workflows, consistent processes, specialized agent delegation
Adoption & trust: 3.7k installs on skills.sh; 2k GitHub stars; 3/3 security scanners passed (skills.sh audits).
What problem does it solve?
You repeat the same multi-step agent instructions in chat instead of a stable, shareable `/command` your agent can execute on demand.
Who is it for?
Solo builders standardizing Claude Code rituals (reviews, scaffolding, deploy prep) as project or global slash commands.
Skip if: One-off tasks you will never run again, or teams that do not use Claude Code slash-command mechanics.
When should I use this skill?
You need to create a new slash command from scratch, automate repetitive workflows, or optimize command structure and location for Claude Code.
What do I get? / Deliverables
You get a well-structured slash command with clear agent instructions, sensible tool usage, and a location strategy—ready to invoke and refine in real sessions.
- Slash command markdown with agent-executable instructions
- Documented command pattern and placement strategy
- Test plan for verifying autonomous command runs
Recommended Skills
Journey fit
Spans multiple journey phases - primary shelf plus alternate fits below.
Slash commands are agent infrastructure you author during product work; the canonical shelf is Build because you define `.claude/commands` (or project commands) alongside the repo. Command Creator is explicitly about agent-executable workflows, tool patterns, and delegation—not app feature code.
Where it fits
Author `/scaffold-api` with fixed steps and tool boundaries before feature work starts.
Encode `/pre-merge-check` so every PR gets the same lint, test, and diff review ritual.
Add `/triage-incident` that walks logs, hypotheses, and rollback steps after an alert fires.
How it compares
Use for authoring durable slash-command playbooks—not for ad-hoc single prompts or MCP server wiring.
Common Questions / FAQ
Who is command-creator for?
Developers and indie builders using Claude Code who want repeatable `/command-name` workflows instead of retyping long instructions every session.
When should I use command-creator?
When creating a new slash command from scratch, automating a repetitive workflow, documenting a consistent process for reuse, or choosing project vs global command placement—during Build agent-tooling setup and again in Ship/Operate when you codify review or ops rituals.
Is command-creator safe to install?
Review the Security Audits panel on this Prism page and inspect the skill bundle before enabling; commands you author may request tools (shell, git, network) depending on what you embed in each slash command.
SKILL.md
READMESKILL.md - Command Creator
# Command Creator A comprehensive skill for creating optimized, agent-executable slash commands in Claude Code. This skill guides you through the entire process of designing, implementing, and testing reusable workflow automation commands. ## Table of Contents - [Overview](#overview) - [When to Use This Skill](#when-to-use-this-skill) - [What Are Slash Commands?](#what-are-slash-commands) - [Key Features](#key-features) - [How It Works](#how-it-works) - [Command Patterns](#command-patterns) - [Location Strategy](#location-strategy) - [Bundled Resources](#bundled-resources) - [Usage Examples](#usage-examples) - [Best Practices](#best-practices) - [Common Use Cases](#common-use-cases) --- ## Overview The Command Creator skill helps you transform repetitive workflows into reusable slash commands that can be invoked with `/command-name` in Claude Code conversations. It provides expert guidance on command structure, agent optimization, and best practices to ensure your commands execute reliably and autonomously. **Purpose**: Create high-quality, agent-executable slash commands with proper structure, clear instructions, and optimal tool usage patterns. **Target Users**: Developers who want to: - Automate repetitive workflows - Document consistent processes for reuse - Create project-specific or global automation - Delegate complex tasks to specialized agents --- ## When to Use This Skill Invoke this skill when you need to: - Create a new slash command from scratch - Automate a workflow you find yourself repeating - Document a multi-step process for consistent execution - Convert manual procedures into automated commands - Create project-specific commands for team workflows - Build global commands for personal productivity **Trigger Phrases**: - "create a command" - "make a slash command" - "add a command" - "I keep doing X, can we make a command for it?" - "automate this workflow" - "create a reusable command" --- ## What Are Slash Commands? Slash commands are markdown files stored in `.claude/commands/` (project-level) or `~/.claude/commands/` (global/user-level) that get expanded into prompts when invoked. **Structure**: ```markdown --- description: Brief description shown in /help (required) argument-hint: <placeholder> (optional, if command takes arguments) --- # Command Title [Detailed instructions for the agent to execute autonomously] ``` **Invocation**: ``` /command-name [arguments] ``` **Storage Locations**: - **Project-level**: `.claude/commands/my-command.md` (only available in this project) - **Global/User-level**: `~/.claude/commands/my-command.md` (available everywhere) --- ## Key Features ### 1. Intelligent Location Detection Automatically determines whether commands should be project-level or global based on: - Current directory git repository status - User explicit preferences - Command scope and purpose ### 2. Pattern-Based Design Guides you through proven command patterns: - **Workflow Automation**: Multi-step processes with analysis, action, and reporting - **Iterative Fixing**: Continuous improvement loops (run → parse → fix → repeat) - **Agent Delegation**: Complex tasks broken into specialized agent work - **Simple Execution**: Direct tool or script execution with arguments ### 3. Agent-Optimized Instructions Creates commands that agents can execute autonomously with: - Imperative/infinitive verb-first instructions - Explicit tool usage specifications - Clear success criteria - Concrete error handling - Expected outcomes defined ### 4. Quality Assurance Includes comprehensive best practices for: - Proper naming conventions (kebab-case enforced) - Argument handling and hints - Tool restriction guidelines - Error recovery strategies - Progress reporting patterns ### 5. Bundled Reference Documentation Provides three comprehensive reference files: - **patterns.md**: Command design patterns with detailed examples - **examples.md**: Real-world command implementations - **best-practic