ClaudeCommandsExplorer
rhassank/ClaudeCommandsExplorer
Searchable Claude Code slash command playbook with examples, workflows, and custom skill templates.
Instalação
npx convez-ai rhassank/ClaudeCommandsExplorer1 comando no terminal: detecta o formato do repositório (marketplace ou plugin avulso) e instala via Claude Code. Requer Node 18+ e o Claude Code instalado.
Métricas
README
A tradução para o português deste README ainda está na fila.
⚡ Claude Code Slash Commands
A searchable Claude Code slash command reference — with examples, workflows, and custom skill templates
A curated reference for built-in commands, bundled skills, and workflows, sourced from Anthropic's official documentation. Plus a complete guide to authoring custom commands as skills.
🖥️ Open the interactive reference → · 📖 Command reference · 🛠️ Build your own · 🔗 Live demo
[!IMPORTANT] Disclaimer: This is an independent, unofficial community resource. It is not affiliated with, endorsed by, or sponsored by Anthropic. See the full Disclaimer below. Everything here is sourced from the official Claude Code documentation — always verify against the live docs, since Claude Code changes often.
👤 Who this is for
This repository is for developers using Claude Code who want a fast, searchable way to understand slash commands, when to use them, and how to build their own custom commands using skills. It is especially useful for:
- Developers new to Claude Code who want a quick command overview
- Teams standardizing Claude Code workflows across a project
- Engineers building reusable custom skills for their codebase
- Maintainers documenting AI-assisted development practices
📑 Table of Contents
- Who this is for
- Disclaimer
- Official documentation
- What slash commands are
- The three kinds of commands
- Quick start
- Browse the commands
- Build your own
- Documentation
- Repository structure
- Contributing
- License
⚠️ Disclaimer
This repository is an independent, unofficial community reference. It is not affiliated with, endorsed by, sponsored by, or maintained by Anthropic. "Claude" and "Claude Code" are trademarks of Anthropic, used here only to describe the tool this reference documents.
- The content is compiled from Anthropic's public documentation and reflects a snapshot in time. Claude Code is updated frequently, so commands may be added, renamed, or removed after this was written.
- This reference may contain errors, omissions, or outdated information. It is provided "as is," without warranty of any kind (see LICENSE).
- The official Anthropic documentation is always the authoritative source. When in doubt, follow the official docs and type
/in your own Claude Code session to see what's actually available to you.
If you represent Anthropic and would like any content changed or removed, please open an issue.
🔗 Official Documentation
Always verify against — and prefer — the official source:
| Topic | Official Anthropic documentation |
|---|---|
| 📋 All commands (built-in, skills, workflows) | https://code.claude.com/docs/en/commands |
| 🧩 Skills & custom commands | https://code.claude.com/docs/en/slash-commands |
| 🔀 Workflows | https://code.claude.com/docs/en/workflows |
| 🧠 Subagents | https://code.claude.com/docs/en/sub-agents |
| 🔌 Plugins | https://code.claude.com/docs/en/plugins |
| 📚 Claude Code docs home | https://code.claude.com/docs |
[!NOTE] Every documentation page in this repo also links back to its specific official source at the bottom.
🤔 What slash commands are
In Claude Code, slash commands control the session from inside it — switch models, manage permissions, clear context, review a diff, run a workflow. Type / to see what's available, or / plus letters to filter.
Two rules: a command is only recognized at the start of your message, and text after the command name is passed to it as arguments.
/review 1234
/review is the command; 1234 is the argument.
🧩 The three kinds of commands
| Kind | What it is | Examples |
|---|---|---|
| Built-in | Behavior coded into the CLI | /init · /clear · /compact · /model · /diff · /review · /security-review |
| Bundled skill | A prompt handed to Claude that it can also trigger automatically | /code-review · /debug · /batch · /loop · /run · /verify · /simplify |
| Bundled workflow | A dynamic workflow that fans work across subagents | /deep-research |
And the fourth, most powerful option: commands you write yourself as skills.
🚀 Quick start
- Browse — open
index.htmlfor the searchable, filterable reference (press / to search, filter by type or workflow group). - Look it up — or read the Command Reference grouped by workflow phase.
- Build your own — turn a repeated prompt into a
/commandwith the authoring guide.
# In Claude Code, try a real bundled skill:
/code-review --fix
For more in-depth examples — /code-review ultra --comment, /batch, /loop, /deep-research, and combined workflows — see Advanced Command Examples.
🗂️ Browse the commands
| Section | What's inside | Open |
|---|---|---|
| 🖥️ Interactive reference | 86+ commands, searchable & filterable | → |
| 📖 Command reference | The full table, grouped by workflow phase | → |
| 🧩 Bundled skills | /code-review, /debug, /batch, /run… | → |
| 🔀 Bundled workflows | /deep-research and the workflow view | → |
[!TIP] In the interactive site, the Skill and Workflow badges mark the prompt-based commands; everything else is built into the CLI.
🛠️ Build your own
Custom commands are skills. Create a SKILL.md and its directory name becomes the command:
# ~/.claude/skills/summarize-changes/SKILL.md -> /summarize-changes
---
description: Summarize uncommitted changes and flag risks.
---
## Current changes
!`git diff HEAD`
## Instructions
Summarize the changes above, then list any risks.
The !`git diff HEAD` line injects the live diff before Claude reads the prompt; $ARGUMENTS (and $0, $1, named args) capture input. Full details:
- Authoring Skills — the complete model, frontmatter, scopes
- Arguments & Injection —
$ARGUMENTSand!`cmd` - Custom Skill Examples — copy-ready skills
- SKILL.md template · command-file template
📖 Documentation
| Guide | What you'll learn |
|---|---|
| Getting Started | How commands work; the three kinds |
| Command Reference | Curated command list, grouped by workflow phase |
| Bundled Skills | The prompt-based commands that ship with Claude Code |
| Bundled Workflows | /deep-research and managing background work |
| Authoring Skills | Build and share your own commands |
| Arguments & Injection | Pass input and inject live data |
| Best Practices | Write skills that stay reliable |
| Advanced Command Examples | Deep dives on /code-review, /batch, /loop, /deep-research, and more |
| Custom Skill Examples | Copy-ready SKILL.md files |
🧱 Repository structure
claude-code-slash-commands/
├── index.html # Searchable interactive reference
├── README.md
├── CONTRIBUTING.md
├── LICENSE
├── docs/
│ ├── getting-started.md
│ ├── command-reference.md # full official table (generated)
│ ├── bundled-skills.md
│ ├── bundled-workflows.md
│ ├── authoring-skills.md
│ ├── arguments-and-injection.md
│ └── best-practices.md
├── examples/
│ ├── custom-skill-examples.md # real, copy-ready SKILL.md files
│ └── team-and-sharing.md
├── templates/
│ ├── skill-template.md
│ └── command-file-template.md
└── assets/
├── images/
└── badges/
🤝 Contributing
Found a command that changed, or want to add a custom-skill example? See CONTRIBUTING.md. The golden rule: everything must match the official docs — cite the source for any command claim.
📄 License
Licensed under the MIT License — use it, fork it, share it. The software and content are provided "as is," without warranty of any kind, as stated in the license and the Disclaimer above.
The MIT license covers this repository's own content only. It does not grant any rights to Anthropic's trademarks or documentation; "Claude" and "Claude Code" remain trademarks of Anthropic.