Agentic Coding with Claude Code : The Everyday Developer's Guide to Agentic Coding with Claude Code.

Format:

Book

Author/Creator:

Marco, Eden.

Language:

English

Subjects (All):

Intelligent agents (Computer software).

Physical Description:

1 online resource (377 pages)

Edition:

1st ed.

Place of Publication:

Birmingham : Packt Publishing, Limited, 2026.

Summary:

Move beyond prompt-based coding by using Claude Code to design reliable, agent-driven workflows and build production-ready full-stack applications Key Features Learn Claude Code as a development partner to build custom automations and agentic workflows Build and iterate on a Next.js application using Claude Code for AI-assisted development Scale.

Contents:

Intro

Agentic Coding with Claude Code

The everyday developer's guide to agentic coding with Claude Code

Contributors

About the author

About the reviewers

Table of Contents

Preface

Who this book is for

What this book covers

To get the most out of this book

Download the example code files

Download the color images

Conventions used

Get in touch

Share your thoughts

Free benefits with your book

How to unlock

Part 1

Foundations of Context Engineering and Claude Code

1

Context Engineering

Introduction to context engineering

From prompt engineering to context engineering

Why context matters for agents

Claude Code and its context engineering strategies

Strategy 1: Writing context and persistent memory

Strategy 2: Intelligent context retrieval

Strategy 3: Compression of context

Strategy 4: Context isolation with subagents

System prompts and why they matter

Best practices for crafting system prompts

The Goldilocks zone for system prompts

Problem with overly specific prompts

Problem with overly vague prompts

What a good middle-ground prompt looks like

Summary

Get this book's PDF copy, code bundle, and more

2

The Gist of Claude Code

Setting up the Next.js project

Running and verifying the application

Initializing Claude code

Contents of the CLAUDE.md file

Connecting Playwright MCP to Claude Code

Verifying the MCP connection

Using Playwright MCP to pen a browser

Using cursor rules as additional context

Creating a frontend memory context

Using pec-driven design

Reviewing the generated spec

Implementing the main page using the SPEC

Running the implementation

3.

Getting Started with Claude Code - A Tour of Essential Commands

Pricing and plans for Claude Code

Option 1: Using an Anthropic API key

Points to keep in mind when using Anthropic API

Important considerations when using an API key

Option 2: Using a Claude subscription

Available models under this option

Subscription plans and pricing

Controlling Claude Code with slash commands

Integrating Claude Code with the IDE

Using hooks in Claude Code

Creating a notification hook

Initializing the environment

Understanding memory in Claude Code

Types of memory

How does Claude Code read memory?

The Claude Code memory hierarchy

How upward discovery works

Lazy loading of nested memory files

Adding and editing memory in Claude Code

Initializing memory with /init

Hands-on example: Using memory with an existing project

Verifying project memory

Adding and using memory entries

Working with multiple memory files

Rewind feature in Claude Code

Benefits and importance of the rewind feature

Using rewind in practice

Custom slash commands in Claude Code (powered by Skills)

Creating a skill

Executing the skill

Improving the skill with ARGUMENTS

Improving auto-generated commit messages with context engineering

Part 2

Extending Claude Code: Protocols, Automation, and Structured Workflows

4

Extending Claude Code with MCP Servers and Plugins

Why MCP?

The integration problem

The portability challenge

MCP as an abstraction layer

Network effects and ecosystem growth

MCP core architecture

Advantages of MCP

Core components of MCP

MCP hosts

MCP servers

MCP clients

MCP in action

Behavior without an MCP server

Connecting an MCP server

Reusing the same MCP server across clients.

Connecting Claude Code to MCP server

Adding Context7 MCP Server via the CLI

MCP server scope and configuration

Restarting Claude Code and enabling the MCP server

Listing and inspecting installed MCP servers

Querying LangGraph using the Context7 MCP server

Persisting MCP server usage in project memory

Verifying persistent MCP behavior

Committing MCP artifacts to version control

MCP scope cheat sheet

When to Use Each MCP Scope

Context engineering in the MCP ecosystem

The problem - overly general MCP configuration

Example repository and MCP server

Running the MCP server

Adding the running MCP server to Claude Code

Project-level MCP configuration

Inspecting context usage

Reducing context by scoping MCP configuration

Enforcing a strict MCP configuration

Managing MCPs dynamically within a session

Current limitations of MCP

Context pollution

Inefficient agent execution

The native tongue problem - JSON versus code

Limitations of tool definitions and schemas

CodeMode by Cloudflare

Key observations

Tool calling and its constraints

The role of MCP

Integrating CodeMode

Converting MCP to TypeScript

Execution in a sandbox

Managing configuration with Claude Code plugins

Benefits of the plugin system

Exploring the official plugin marketplace

Adding a marketplace to Claude Code

Reviewing plugin sources and security considerations

Installing and managing plugins

Installing a plugin

Verifying the installation

Updating a plugin

Attempting to uninstall a plugin

Using the feature-dev plugin

Starting the workflow

Handling Git state during the feature-dev workflow

Checking the Git state

Stashing local changes

Switching branches and restoring state

Providing feature details

Inspecting the README file in the feature branch.

Switching back to the main branch

Reviewing and refining the proposed change

Committing, pushing, and verifying the change

Plugin marketplaces and enterprise use

5

Automating Your Development Workflow with Claude Code and GitHub

Installing and configuring the Claude Code GitHub integration

Installing prerequisites

Installing the Claude GitHub app

Installing the GitHub workflow

Creating the pull request

Using Claude Code to resolve GitHub issues

Claude's to-do list and context discovery

Reviewing the proposed changes

Understanding Claude's workflow

Adding repository context with CLAUDE.md

Initializing the repository context

Reviewing and committing CLAUDE.md

Inspecting Claude's changes

6

Claude Code Planning and Multi-agent Workflows

Claude Code planning mode

Characteristics of planning mode

Planning workflow

Benefits of spec-driven development

Example: Iterating on the hook marketplace specification

Persisting the hook marketplace specification

Using multiple Claude Code agents in parallel

Identifying suitable tasks for parallel execution

Example: Parallel development in the HookHub project

Assigning independent tasks to agents

Executing tasks concurrently

Reviewing and committing changes

7

Working with Claude Code Subagents

Introduction to Claude Code subagents

Subagent file structure

Relationship to React Agents

Configuring the first subagent

Defining the agent description

Applying the system prompt

Selecting tools for the agent

Choosing the model and visual settings

Reviewing the generated agent configuration

Testing our subagent.

Running a code review subagent

Spawning multiple review instances

Context flow while using subagents

Context windows in action

Example: Creating a Mermaid diagram subagent

Creating subagent configuration

Tool selection and context considerations

Understanding the subagent configuration

Invoking the subagent

Evaluating the results

Refining the subagent description

Observing the updated prompt flow

Evaluating the subagent output

Refining the sub-agent output

Observing the updated execution flow

Scaling with concurrent subagents

Setting up the environment

Installing dependencies

The infinite slash command

Reorganizing the hero component

Generating the spec file from the Hero file

Running the infinite slash command

Cleaning the context before refinement

What we have done so far

Infinite prompt

Phase 1: Specification analysis

Phase 2: Output directory reconnaissance

Phase 3: Iteration strategy

Phase 4: Parallel agent coordination

Agent assignment protocol

Subagent task structure

Parallel execution management

8

Creating and Customizing Output Styles

Creating a custom output style with Claude Code

Creating and reviewing a new output style

How output styles work

Why YAML

Creating YAML as an output style

Creating an advanced project-level output style

Defining a custom HTML output style

Using the HTML output style

Automating file generation

Working with multiple output styles

Launching a new Claude Code instance

Opening a second instance with a different output style

Parallel output styles in practice

Specialized output style workflows

Customizing the status line in Claude Code

Defining a custom status line command

Initializing the status line setup.

Updating the configuration.

Notes:

Description based on publisher supplied metadata and other sources.

ISBN:

1-80602-258-3

9781806022588

OCLC:

1582440253

Publisher Number:

CIPO000361102