daBongo LMS AI Training Courses

← Back to Course

Claude Code Foundations – Your Daily Dev Workflow

Lesson 2: Setting Up and Configuring Claude Code

Lesson Objectives

By the end of this lesson, students should be able to:

  • Complete the Claude Code installation and authentication process
  • Configure IDE integration for their editor
  • Write a functional first CLAUDE.md for a real project
  • Understand the key settings that affect Claude Code's behavior

Lesson Content

Installation.

Claude Code is installed as a CLI tool. Current installation instructions are at docs.anthropic.com/en/docs/claude-code/setup – always follow the current documentation rather than memorized steps, as the installation process may change between versions.

General setup flow:

  1. Install via the documented package manager (npm global install as of 2026 – verify current method)
  2. Run claude from your terminal to launch
  3. Authenticate with your Anthropic account or API key
  4. Verify the installation by running a quick test session

Authentication.

Claude Code uses your Anthropic account or API key for authentication. For personal use, Claude.ai account authentication is the standard path. For team use or programmatic integration, API key authentication may be appropriate. Verify current authentication options at docs.anthropic.com.

IDE integration.

Claude Code has IDE extensions for VS Code and JetBrains IDEs (verify current availability). The extension surfaces Claude Code within your editor – providing an inline panel rather than requiring a separate terminal window. For terminal-first developers, the standalone terminal mode is fully featured. For IDE-centric workflows, the extension reduces context switching.

Key configuration settings.

Claude Code has configurable settings (verify current settings reference at docs.anthropic.com/en/docs/claude-code/settings):

  • Auto-approve permissions: Which types of actions Claude can take without prompting – file reads, file edits, shell commands. Start with conservative settings and expand as you build trust with the tool in your workflow.
  • Model selection: Which Claude model to use. Higher capability models produce better results but consume more tokens. Configure to match your use case and budget.

Writing your first CLAUDE.md.

A minimal effective CLAUDE.md covers: “`markdown

Project: [Name]

Tech Stack

[Language, framework, database, key libraries]

Architecture Notes

[Brief description of the main components and how they relate]

Conventions

[Naming conventions, code style, patterns to follow]

Current Focus

[What is actively being developed right now]

Test Command

[How to run tests: npm test, pytest, etc.]

Things to Avoid

[Anti-patterns, deprecated approaches, things Claude should not do here] “`

Keep it short and current. A CLAUDE.md that accurately reflects the project is far more valuable than a comprehensive one that is six months out of date.

Practical Example

A developer joins a new team and gets a codebase access.

Before writing a line of code, he spends thirty minutes creating a CLAUDE.md by interviewing the codebase: he asks Claude Code to scan the project structure and summarize it, then uses that summary as the starting point for his CLAUDE.md.

He adds the team's test command, the naming conventions from the existing code, and the three architectural decisions the team lead explained to him during onboarding.

His Claude Code sessions are immediately more targeted because Claude has project context from day one.

Safety Notes

When configuring auto-approve permissions, start restrictive and expand deliberately. Allowing Claude to auto-approve shell command execution without understanding what commands it may run is a significant permission grant. Review the current permissions model documentation carefully before enabling broad auto-approval, especially in directories containing sensitive data or production configuration.

Log in and enroll to access lesson quizzes.

← Back to Course

Scroll to Top