Claude Code is Anthropic's agentic coding tool for the terminal, designed as an autonomous agent rather than a suggestion engine. It reads files, writes code across multiple files, executes shell commands, works with Git, and connects to external services through the Model Context Protocol (MCP). If you're comfortable in the terminal, but new to agentic AI tools, this book is your starting point. You'll learn what the documentation doesn't cover: where Claude Code looks for CLAUDE.md files, how skill path resolution works, and when headless mode beats an interactive session.
In Claude Code: Up and Running, educator and Java Champion Ken Kousen takes you through hands-on examples covering permissions, trust levels, context windows, MCP integrations, model switching, custom slash commands, and multi-agent workflows. A companion GitHub repository with CLAUDE.md templates, hooks, and MCP snippets remains a useful resource long after your first read.

• Install and authenticate Claude Code within 15 minutes
• Configure CLAUDE.md files and project-specific behavior
• Connect external services through MCP integrations
• Work with permissions, trust levels, and context windows
• Run Claude Code in terminal, IDE, and CI/CD environments