Claude Code Visual Tutorial: 10 Modules, Mermaid Diagrams, and Ready‑to‑Use Templates

Overview

The Claude HowTo repository provides a structured, visual, example‑driven tutorial for Claude Code. It shows how to use every Claude Code capability and supplies production‑ready templates so that users can immediately apply the tools in real projects.

Core Features Compared to the Official Documentation

Format : Official docs are plain reference; Claude HowTo adds Mermaid flowcharts for visual guidance.

Depth : Official docs list functions; Claude HowTo explains the underlying mechanisms.

Examples : Official docs give basic snippets; Claude HowTo provides ready‑to‑use production templates.

Structure : Official docs are organized by function; Claude HowTo offers a progressive learning path from beginner to advanced.

Onboarding : Official docs assume self‑learning; Claude HowTo includes a guided roadmap with time estimates.

Self‑assessment : Not present in official docs; Claude HowTo includes interactive quizzes ( /self-assessment or /lesson-quiz) to identify knowledge gaps.

Tutorial Contents

✅ 10 tutorial modules covering every Claude Code feature, from slash commands to custom agents.

✅ Copy‑and‑paste configurations for slash commands, CLAUDE.md memory files, hook scripts, MCP settings, sub‑agent definitions and full plugin packs.

✅ Mermaid flowcharts that illustrate the internal workings of each feature.

✅ Guided learning path of 11–13 hours from entry‑level to mastery.

✅ Built‑in self‑assessment commands ( /self-assessment, /lesson-quiz) that highlight blind spots.

Learning Path (Modules & Estimated Time)

Slash Commands – Beginner – 30 min

Memory – Beginner+ – 45 min

Checkpoints – Intermediate – 45 min

CLI Basics – Beginner+ – 30 min

Skills – Intermediate – 1 h

Hooks – Intermediate – 1 h

MCP – Intermediate+ – 1 h

Sub‑Agents – Intermediate+ – 1.5 h

Advanced Features – Advanced – 2–3 h

Plugins – Advanced – 2 h

Total: 11–13 hours to complete the full path.

15‑Minute Quick Start

Step 1 – Clone the Guide

git clone https://github.com/luongnv89/claude-howto.git
cd claude-howto

Step 2 – Copy the First Slash Command

mkdir -p /path/to/your-project/.claude/commands
cp 01-slash-commands/optimize.md /path/to/your-project/.claude/commands/

Step 3 – Try It

In Claude Code type:

/optimize

Step 4 – Set Project Memory

cp 02-memory/project-CLAUDE.md /path/to/project/CLAUDE.md

Step 5 – Install a Skill

cp -r 03-skills/code-review ~/.claude/skills/

Core Feature Details

1. Slash Commands – Custom Shortcuts

Definition : Markdown files stored under .claude/commands/, each representing a pre‑defined high‑level instruction.

Purpose : Eliminates the need to type long prompts repeatedly.

Usage : Define commands such as /optimize (focus on performance) or /pr (generate a pull‑request summary). Execute with a single keystroke.

Location : 01-slash-commands/ Examples : optimize.md – code‑optimization analysis pr.md – pull‑request preparation generate-api-docs.md – API documentation generator

Install :

cp 01-slash-commands/*.md /path/to/project/.claude/commands/

Run :

/optimize
/pr
/generate-api-docs

2. Memory – Persistent Context

Definition : CLAUDE.md files placed at the project root or specific directories.

Purpose : Keeps Claude aware of project conventions, build commands, language versions, etc.

Usage : Acts as an "onboarding guide" that Claude reads on startup, automatically loading Go version, lint rules, test commands, etc.

Location : 02-memory/ Examples : project-CLAUDE.md – team‑wide standards directory-api-CLAUDE.md – directory‑specific rules personal-CLAUDE.md – personal preferences

Install :

# Project memory
cp 02-memory/project-CLAUDE.md /path/to/project/CLAUDE.md
# Directory memory
cp 02-memory/directory-api-CLAUDE.md /path/to/project/src/api/CLAUDE.md
# Personal memory
cp 02-memory/personal-CLAUDE.md ~/.claude/CLAUDE.md

Run : Loaded automatically when Claude starts.

3. Skills – Automated "Brain Extensions"

Definition : Complex logic packages (commands + optional scripts) stored under ~/.claude/skills/.

Purpose : Executes recurring logic without manual prompting.

Usage : A "code‑review" skill runs automatically after Claude generates code, performing background analysis.

Location : 03-skills/ Examples : code-review/ – comprehensive code review with scripts brand-voice/ – brand‑tone consistency checker doc-generator/ – API documentation generator

Install :

# Personal skill
cp -r 03-skills/code-review ~/.claude/skills/
# Project skill
cp -r 03-skills/code-review /path/to/project/.claude/skills/

Run : Invoked automatically when relevant.

4. Sub‑Agents – Specialized Virtual Teams

Definition : Small AI instances with their own role prompt and isolated permissions.

Purpose : Allows Claude to act as distinct experts (e.g., architect, security specialist, junior developer) without role conflict.

Usage : Call a sub‑agent with @test-engineer to get dedicated testing advice.

Location : 04-subagents/ Examples : code-reviewer.md – comprehensive code quality analysis test-engineer.md – testing strategy and coverage documentation-writer.md – technical documentation secure-reviewer.md – read‑only security review implementation-agent.md – full‑function implementation

Install :

cp 04-subagents/*.md /path/to/project/.claude/agents/

5. MCP (Model Context Protocol) – Connecting AI to the Outside World

Definition : Standardized interface that lets Claude access external tools, databases and APIs.

Purpose : Overcomes the limitation that the model cannot see GitHub issues, database schemas, or live logs.

Usage : With MCP, Claude can query Jira tickets or read production logs directly.

Location : 05-mcp/ Install :

export GITHUB_TOKEN="token"
claude mcp add github -- npx -y @modelcontextprotocol/server-github

6. Hooks – Event‑Driven Automation

Definition : Scripts triggered on specific events (session start, before/after command execution).

Purpose : Integrates Claude into CI/CD pipelines or linting workflows.

Usage : A hook can run git status or a Lint check after Claude modifies code, achieving unattended programming.

Location : 06-hooks/ Install :

mkdir -p ~/.claude/hooks
cp 06-hooks/*.sh ~/.claude/hooks/
chmod +x ~/.claude/hooks/*.sh

7. Plugins – Packaged Feature Collections

Definition : Distributable bundles that include commands, skills and rules.

Purpose : Provides a one‑click way to install the full set of nine core modules for teammates.

Usage : Installing everything-claude-code gives you all configurations without manual copying.

Location : 07-plugins/ Install :

/plugin install pr-review

8. Checkpoints – “Undo” Points for Programming

Definition : Git or file‑system snapshots taken during a session.

Purpose : Allows instant rollback if Claude’s changes break the code.

Usage : Claude creates a checkpoint before large refactors; you can revert to the previous state on failure.

Location : 08-checkpoints/ Run : Automatically enabled; can be configured in settings.

9. Advanced Features – Planning and Thinking Modes

Definition : Modes such as "Planning Mode" and "Thinking Process" that make Claude outline a plan before writing code.

Purpose : Prevents the model from jumping straight to code on complex problems, reducing logical errors.

Usage : In Planning Mode Claude presents a modification plan for user confirmation before execution, acting as a safety net for large refactors.

Location :

09-advanced-features/

10. CLI Basics – Command‑Line Control

Definition : Terminal flags and interactive options.

Purpose : Enables efficient startup and control of the Claude environment.

Usage : Use --non-interactive for scripting or --print-config to debug configuration.

Location :

10-cli/

Typical Use Cases

Automated code review – combine slash commands, sub‑agents, memory and MCP.

Team onboarding – memory + slash commands + plugins.

CI/CD automation – CLI reference + hooks + background tasks.

Documentation generation – skills + sub‑agents + plugins.

Security audit – sub‑agents + skills + read‑only hooks.

DevOps pipelines – plugins + MCP + hooks + background tasks.

Complex refactoring – checkpoints + planning mode + hooks.

Reference Resources

Claude Code official docs: https://docs.anthropic.com/en/docs/claude-code
GitHub repository: https://github.com/luongnv89/claude-howto