Learn Claude Code
19 lessons across five levels, about 2.5 hours in total. Work through them in order, or jump to the level that matches you.
Level 1: Curious
What Claude Code is, how to install it and how to ask for what you want.
What is Claude Code? Agentic coding explained
A beginner's guide to Claude Code: what agentic coding means, how it differs from a chatbot, and what you can build with it.
Every way to use Claude Code
Terminal, VS Code, JetBrains, desktop, web, phone, GitHub, Slack, scripts and the Agent SDK: every place Claude Code runs and when to use each one.
Install Claude Code and start a session
Step-by-step: install Claude Code on Mac, Windows or Linux, log in, open a project folder and run your first agentic coding session with confidence.
How to talk to Claude Code: better prompts
Write clear Claude Code prompts: give context, define done, share examples and break big jobs into steps for better results with fewer retries.
Level 2: Getting going
Editing real files safely, steering with modes and undoing mistakes.
How Claude Code works under the hood
What happens after you press Enter: the agentic loop, the tools Claude uses, how context and memory work, and where permissions and checkpoints fit.
Reading and editing files, plus @ mentions
How Claude Code reads, searches and edits your files, how to review diffs, and how to use @ mentions to point Claude at the right files and folders.
Claude Code permissions and plan mode
Claude Code permission prompts, auto-accept edits and plan mode: when to use each, how Shift+Tab switches between them, and how to stay safe.
Git, undo and sessions: rewind and resume
Make mistakes safely in Claude Code. Use git commits, checkpoints and rewind to undo changes, and claude -c or -r to continue or resume previous sessions.
What can you use Claude Code for? 12 everyday jobs
Twelve practical ways to use Claude Code: understanding a codebase, fixing bugs, writing tests, reviewing pull requests and building internal tools.
Level 3: Daily driver
Memory, context and custom commands that make Claude feel like a teammate.
CLAUDE.md explained: memory and /init
How CLAUDE.md memory files work in Claude Code, how /init creates one, the user, project and folder hierarchy, and what to put in yours.
Slash commands and context management
Master Claude Code's built-in slash commands and the context window: when to /clear, when to /compact, and how to keep long sessions sharp.
Custom slash commands and settings.json
Create your own Claude Code slash commands with markdown files, and configure settings.json for permissions, environment and team-wide defaults.
Level 4: Power user
Skills, subagents, hooks and MCP: extend Claude Code to fit your workflow.
Claude Code skills and subagents
Learn how Claude Code skills (SKILL.md) package reusable know-how and how subagents work in isolated context to keep your main session clean and fast.
Claude Code hooks: automatic guardrails
Use Claude Code hooks to run your own scripts at key moments: auto-format after edits, block dangerous commands, add context and enforce rules every time.
Does Claude Code write bloated or insecure code?
Why AI-written code can be bloated, chatty or insecure, and the CLAUDE.md rules, hooks and reviews that keep Claude Code output lean and safe.
MCP servers and plugins for Claude Code
What the Model Context Protocol (MCP) is, how to add MCP servers to Claude Code for GitHub, databases and docs, and how plugins bundle it all together.
Level 5: Pro
Automation, the Agent SDK and architecting production-grade agentic systems.
Headless Claude Code and CI/CD pipelines
Run Claude Code non-interactively with -p, get JSON output for scripts, and automate pull request reviews in CI/CD with the Claude Code GitHub Action.
The Claude Agent SDK: build your own agents
Use the Claude Agent SDK in Python or TypeScript to build custom agents on the engine behind Claude Code, with tools, permissions, hooks and MCP built in.
Architecting agentic systems
Design production multi-agent systems with Claude: coordinators, context passing, tool design, guardrails and human-in-the-loop. Architect exam prep.