Last verified
Learn Claude Code through 10 structured tutorial modules covering slash commands, memory, hooks, subagents, MCP, and plugins, with copy-paste templates.
Start with the Beginner track: slash commands (30 min) through CLI basics (30 min). Copy the first slash command template into your project, try /optimize, and have a working Claude Code customization in 15 minutes. The self-assessment quiz surfaces which modules matter most for your current use.
If you need video instruction or instructor-led pacing, a paid video course may be a better fit alongside this written guide.
Clone the repository and adapt the hook scripts in 06-hooks/ for your team's CI/CD events: auto-format code after file writes, run tests before commits, send notifications on deploy events. Memory templates in 02-memory/ provide a starting CLAUDE.md with team-wide standards that every team member's Claude Code session loads automatically.
If your team has no current Claude Code users, distributing and maintaining shared templates may not justify the overhead until adoption grows.
Modules 7 (MCP) and 8 (subagents) provide config examples and production templates for common integrations: GitHub MCP, database queries, filesystem access, and multi-server configs. The subagent definitions in 04-subagents/ cover code review, test engineering, documentation writing, and security audits.
If you are building MCP servers rather than consuming them, this guide covers the client side only and does not address server development.
Run uv run scripts/build_epub.py to generate a complete EPUB with all modules and rendered diagrams. Read on a tablet, phone, or e-reader without network access. The EPUB is generated locally from the cloned repository, so it reflects whatever version you have checked out.
If you prefer video instruction, the EPUB format covers written tutorials only and does not include video content.
Claude Code ships with feature reference documentation but no structured path for learning how to combine those features into real workflows. Developers who install it can run basic prompts but quickly hit a ceiling: they know slash commands exist but not how to chain them with hooks, memory, and subagents into an automated code review or deployment pipeline. Without a clear learning order, most developers skim everything and master nothing.
The result is that most Claude Code users leave the majority of the tool's capabilities unused. Building a pipeline that uses memory to preserve context, delegates tasks to specialized subagents, and triggers security scans automatically via hooks requires knowledge spread across multiple unconnected reference pages. Most developers either spend hours piecing this together or fall back to simple one-shot prompts.
Covers every Claude Code feature in a structured order: slash commands (30 min), memory (45 min), checkpoints (45 min), CLI basics (30 min), skills (1 hr), hooks (1 hr), MCP (1 hr), subagents (1.5 hrs), advanced features (2-3 hrs), and plugins (2 hrs). Each module builds on the previous one, with time estimates so you can plan around your schedule. The full path runs 11-13 hours.
Each module ships ready-to-use files: slash command Markdown files for .claude/commands/, CLAUDE.md templates for project memory, hook scripts for .claude/hooks/, MCP config JSON, and subagent definitions for .claude/agents/. Copy any file into your project and start using it immediately without reading the full module first.
Every feature chapter includes Mermaid diagrams showing how it works internally, not just how to invoke it. Understanding the event order for hooks, how subagent context isolation works, or how MCP tools are selected helps you debug broken workflows and design more reliable automations.
Run /lesson-quiz [topic] directly in Claude Code after each module to check your understanding. The quiz pinpoints gaps so you can revisit specific sections. Run /self-assessment at any time to generate a personalized roadmap based on what you already know and which modules to prioritize.
Run uv run scripts/build_epub.py to generate a full EPUB ebook with all 10 modules and rendered Mermaid diagrams. Useful for reading on mobile devices or without an internet connection. All copy-paste templates are included in context within the ebook.
Both are free resources for learning Claude Code. The official documentation from Anthropic covers every Claude Code feature in depth, organized around individual features. Claude Howto is a tutorial organized as a progressive learning path with Mermaid diagrams, copy-paste templates, and built-in quizzes.
| Feature | Claude Howto | Official Docs |
|---|---|---|
| Format | Progressive tutorial | Feature reference |
| Learning order | Guided 10-module path | Self-directed |
| Internal diagrams | Mermaid architecture diagrams per feature | Minimal |
| Copy-paste templates | Production-ready configs per module | Basic snippets |
| Self-assessment | /lesson-quiz and /self-assessment | None |
The official docs are the authoritative source for exact feature behavior and edge cases. Claude Howto fills the gap between having Claude Code installed and knowing how to build production workflows. Most developers use both: this guide for the learning path, the official docs for specific reference.
Paid courses on platforms like Udemy and Pluralsight offer video instruction for AI coding tools. Claude Howto is MIT licensed, free, and updated with every Claude Code release as a version-controlled repository.
| Feature | Claude Howto | Paid Video Courses |
|---|---|---|
| Cost | Free (MIT) | Paid, per course |
| Format | Written modules with templates | Video lectures |
| Update cadence | Every Claude Code release | Varies by instructor |
| Copy-paste templates | Production-ready configs included | Rarely provided |
| Self-assessment | Built-in Claude Code quizzes | Varies by platform |
Claude Howto is the better choice when you need production-ready templates to copy into your project, or when staying current with the latest Claude Code features matters. Paid video courses are worth considering if you prefer instructor-led pacing or video walkthroughs over written tutorials.
Install the guide by cloning the repository and copying templates into your .claude/ directory.
```bash
git clone https://github.com/luongnv89/claude-howto.git
cd claude-howto
cp 01-slash-commands/*.md /path/to/your-project/.claude/commands/
cp 02-memory/project-CLAUDE.md /path/to/your-project/CLAUDE.md
cp -r 03-skills/code-review-specialist ~/.claude/skills/
```Yes. The guide is MIT licensed and free for personal, commercial, and team use with no restrictions beyond including the license notice. There is no paid tier, no account required, and no usage limits. Clone it, fork it, and adapt it as needed.
The official docs are a feature reference organized around individual features. Claude Howto is a tutorial organized as a progressive learning path, with Mermaid diagrams explaining how each feature works internally and copy-paste templates for immediate use. The two complement each other: use this guide to learn the progression, then use the official docs for exact feature details.
All templates are compatible with Claude Fable 5.1, Opus 5.5, Sonnet 5.5, Haiku 5.5, Sonnet 4.6, Haiku 4.5, and earlier Claude 5 and Claude 4 models. Model-specific notes appear in modules where behavior differs between tiers.
The full 10-module path takes 11-13 hours. You can get immediate value in 15 minutes by copying a slash command template and running it in Claude Code. The three beginner modules (slash commands, memory, CLI basics) total about 1 hour 45 minutes. The advanced modules (subagents, advanced features, plugins) total 5-6 hours.
Yes. Run uv run scripts/build_epub.py from the cloned repository to generate an EPUB ebook with all 10 modules and rendered diagrams. You need uv (the Python package manager) installed. The resulting EPUB includes all copy-paste templates in context.
Alternative to
RevEng.AI
AI-powered reverse engineering for binaries, apps, and assemblies.
Alternative toClaude Managed Agents
Open-source agent deployment API. Self-host or use Omnara Cloud.
Alternative to
Grok Bot
AI teammates you own: your keys, your model, your machine.
Alternative to
Firecrawl
Give any AI agent internet access, free and local
Alternative toGrok Bot
Self-hosted AI agent: persistent memory, multi-channel chat, model choice.
Alternative to
Braintrust
Automated LLM evals, red teaming, and AI vulnerability scanning.
Featured
Alternative toClaude Managed Agents
Open-source agent deployment API. Self-host or use Omnara Cloud.
Alternative to
Superhuman
AI email assistant that drafts replies and clears your inbox
Alternative toGrok Bot
AI teammates you own: your keys, your model, your machine.
The projects we add and every new guide. Free, unsubscribe any time.
Claude Code to 
OpenCode
Jira to 
Windshift
Notion to 
AppFlowy
Slack to 
Mattermost
Linear to 
GitLab
Wispr Flow to 
Handy
CapCut to 
Concat
Grok Bot to 
rakazo
Superhuman to 
Inbox Zero
Airtable to 
AITable
Muse to 
OpenClaw
GitHub Copilot to 
Cline
Datadog to 
Netdata
Microsoft Teams to 
Rocket.ChatConfluence to 
Docmost
Pinecone to 
Qdrant
Zapier to 
n8n
Devin AI to 
OpenHands
Firebase to 
Supabase
1Password to 
KeePassXC
Contentful to 
Strapi
Adobe Premiere Pro to 
Shotcut
ChatGPT to 
AnythingLLM
LangSmith to 
Langfuse
