Module 02 — Claude Code fundamentals

Name: Harness Engineering — Claude Code Mastery
Price: 89 EUR
Availability: InStock

In this module

You will be able to install Claude Code, navigate its terminal interface, understand its agentic loop, write an effective CLAUDE.md, and master plan mode to structure your work before coding.

Contents

2.1 — Installation and first launch Installation methods First steps Key takeaways

2.2 — The terminal interface Screen zones Permission system Key takeaways 2.3 — The agentic loop The cycle in action Practical consequences Key takeaways 2.4 — Built-in tools Read tools (authorized by default) Write tools (permission required) Execution and delegation Limits to know Key takeaways 2.5 — CLAUDE.md: persistent memory The 4 levels Recommended content Importing external files Key takeaways 2.6 — Conditional rules Structure Recommended organization Key takeaways 2.7 — Providing context Explicit context methods Automatic context Key takeaways 2.8 — Plan Mode vs Normal Mode Switching Recommended workflow for non-trivial tasks When to use what Key takeaways 2.9 — Sessions and continuity Context window management Named sessions and resume Headless mode and chaining Key takeaways 2.10 — Built-in git Built-in git commands Safety protocol Git as a source of context Key takeaways 2.11 — Essential keyboard shortcuts The fundamental trio Navigation and editing Slash commands Key takeaways 2.12 — First guided project Step 1 — Setup Step 2 — Configuration Step 3 — Exploration (Plan Mode) Step 4 — Planning (Plan Mode) Step 5 — Implementation (Normal Mode) Step 6 — Verification and commit Key takeaways

2.1 — Installation and first launch Installation methods First steps Key takeaways

2.1 — Installation and first launch

Claude Code is an autonomous terminal agent, not just another IDE extension. It lives in your shell, understands your project, and performs concrete actions on your code. Before anything else, you need to install it — it takes less than a minute.

Installation methods

The recommended method on macOS, Linux, and WSL:

curl -fsSL https://claude.ai/install.sh | bash

Homebrew alternative on macOS:

brew install --cask claude-code

On Windows, two options:

# PowerShell
irm https://claude.ai/install.ps1 | iex

# CMD (requires Git for Windows)
curl -fsSL https://claude.ai/install.cmd -o install.cmd && install.cmd && del install.cmd

Winget is also available:

winget install Anthropic.ClaudeCode

First steps

cd /your/project
claude                           # launches an interactive session
claude "explain this project"    # launches with a direct question
claude -p "query"                # non-interactive mode (print mode), for scripts
claude --continue                # resumes the last conversation
claude --resume                  # resumes a specific session
claude -n "feature-auth"         # named session

Claude Code is tied to your working directory. It detects the stack, the structure, the configuration files. Always navigate to the right folder before launching it.

Key takeaways

• One-command install: curl -fsSL https://claude.ai/install.sh | bash

• First launch: cd /your/project && claude

• Non-interactive mode for scripts: claude -p "query"

• Session resume: --continue (last) or --resume (specific)

• Claude Code is contextual: it understands the project you're in

In this module

2.1 — Installation and first launch Available
2.2 — The terminal interface
2.3 — The agentic loop
2.4 — Built-in tools
2.5 — CLAUDE.md: persistent memory
2.6 — Conditional rules
2.7 — Providing context
2.8 — Plan Mode vs Normal Mode
2.9 — Sessions and continuity
2.10 — Built-in git
2.11 — Essential keyboard shortcuts
2.12 — First guided project

Locked content

Unlock the full course to access this module.

Unlock

Already purchased? Recover access