What you will learn in Claude at Work
Effective AI software development with Claude Code and Claude Cowork
Chapter by chapter
Chapter 1: Meet Claude
The introduction presents Claude as a powerful but context-dependent assistant, argues for critical collaboration, and explains why the author chose it among AI tools.
What you learn in chapter 1 →Chapter 2: Your side of the conversation
The chapter introduction explains how working with Claude across different tasks builds confidence and identifies two foundations for using it productively: understanding the tool and understanding how prompts shape responses.
What you learn in chapter 2 →Chapter 3: Using Claude Desktop and Cowork
The chapter introduction explains why moving from Claude on the web to a permissioned local workspace reduces file transfers and supports Git-backed work.
What you learn in chapter 3 →Chapter 4: Maximizing productivity with skills and schedules
This introduction explains why repeated prompting becomes tedious and presents skills as reusable, adaptable workflows for Claude.
What you learn in chapter 4 →Chapter 5: Getting started with Claude Code
Introduces Claude Code’s ability to work autonomously in a repository and the judgment required to decide how much independence to give it.
What you learn in chapter 5 →Chapter 6: Planning and building in stages with BDD
The chapter introduction explains when to plan with Claude before coding and how plans at different levels make large work easier to review.
What you learn in chapter 6 →Concepts explained in this book
- agent skillA skill is a reusable set of instructions and supporting resources that guides Claude through a particular task or workflow.
- ClaudeClaude is Anthropic's AI assistant and family of large language models.
- behavior-driven developmentBehavior-driven development is a software development technique that specifies and validates behavior through examples expressed in a shared natural-language format.
- CoworkCowork is a Claude Desktop feature that provides a permissioned sandbox for running tools with access to local files and network services.
- promptA prompt is an input message that specifies a question, instruction, or requested task for an AI system.
- implementation planAn implementation plan is a structured description of the work, decisions, and sequence required to build a software feature or system change.
- sandboxingA sandbox environment is an isolated execution environment that restricts code and tool access from the surrounding system.
- Claude CodeClaude Code is a software development tool that assists with writing, running, testing, and debugging code through natural-language interaction.
- contextContext is information supplied to a language model that frames and conditions its response, including relevant prior conversation and background details.
- permissions modelA permission model is a set of rules governing which actions and resource accesses are allowed, require approval, or are blocked.
- artifactA Claude artifact is a standalone generated document or other content item that can be viewed or downloaded separately from a chat.
- subagentA subagent is an agent assigned work by another agent, typically operating in a separate context.
- minimum viable productA minimum viable product is an early implementation containing enough functionality to validate workflows and provide useful value.
- critical dialogueCritical dialogue is an interaction technique that actively questions and evaluates an AI system's suggestions rather than accepting them uncritically.
- connectorA connector is Claude's integration for authenticating to and performing permitted operations on an external service.
- AI agentAn AI agent is a software system that uses AI to pursue tasks through autonomous or semi-autonomous actions.
- strategyA strategy plan is a high-level plan describing the overall approach for delivering a software feature or initiative.
- Gherkin testA Gherkin test is a behavior-focused acceptance test expressed as a structured natural-language scenario.
- delivery planA delivery plan is a staged plan that organizes work into manageable increments for producing and reviewing a deliverable.
- large language modelA large language model is a machine-learning model trained on extensive text data to generate and interpret natural language.
- vertical sliceA vertical slice is an end-to-end implementation that crosses the application's layers or components to deliver a complete feature path.
- training datasetTraining data is the collection of examples used to optimize a machine-learning model's parameters.
- AI modelA model is a trained computational system that processes inputs to produce outputs according to learned patterns.
- generative artificial intelligenceGenerative AI is artificial intelligence that produces new content, such as text, images, or code, from learned patterns.
- Git repositoryA Git repository is a directory and version-control database that stores files, commits, and their change history.
- authorizationAuthorization is the process of granting or restricting a verified user or client permission to perform specified operations.
- action-by-action controlAction-by-action control is a permission technique that lets a user allow, review, or block each external operation individually.
- MCP serverAn MCP server is lightweight middleware that exposes an external system's operations and data to an AI client through the Model Context Protocol.
- MCPThe Model Context Protocol is an open standard for connecting AI models to external systems through discoverable, permissioned operations.
- Docker ComposeA Docker Compose file is a declarative configuration file describing application services, dependencies, and their runtime settings.
- tokenA token is a unit of text processed by a language model, typically representing part of a word, a word, or punctuation.
- cowork projectA Cowork Project is a Claude Desktop project configuration that gives Claude access to a selected local folder and its contents.
- permission modeA permission mode is a Claude Code setting that determines how Claude requests or receives authorization for actions.
- context windowA context window is the maximum amount of tokenized input and conversational context a model can process in one interaction.
- auto permission modeAuto permission mode is a Claude Code mode that uses an AI classifier to evaluate tool permissions instead of repeatedly requiring direct human approval.
- end-to-end testingAn end-to-end test verifies an application flow across its user interface, services, and persistent data.
- goal commandThe /goal command instructs Claude Code to continue working until a specified condition is reached.
- brainstormingA brainstorming technique explores an idea, investigates implementation approaches, and produces a design for implementation.
- REST ClientREST Client is a Visual Studio Code extension for documenting and executing HTTP requests.
- Superpowers pluginSuperpowers is a community Claude Code plugin that provides reusable coding skills.
- diffA diff is a representation of the additions, deletions, and modifications between two versions of content.
- pluginA plugin is a software component that extends a host application's functionality through an established extension interface.
- quality gateA quality gate is a predefined verification criterion that software must satisfy before progressing in a delivery process.
- architecture decision recordAn architecture decision record is a document that captures a significant software architecture decision, its rationale, consequences, and alternatives.
- feature fileA Gherkin feature file is a text file containing structured behavior scenarios for an application feature.
- agentic developmentAgentic coding is a software development approach in which an AI agent directly navigates, modifies, builds, deploys, and tests a repository's codebase.
- linguistic registerRegister is the informal set of language and interaction conventions used when addressing a particular person or audience.
- CLAUDE.mdA CLAUDE.md file is a Markdown instruction file that provides Claude Code with project or personal guidance for software development work.
- dialogueA conversation is a cumulative interaction session consisting of messages exchanged between a user and an AI assistant.
- tool useTool use is an AI technique in which a model invokes external capabilities such as web search or script execution.
- scheduled runA scheduled run is one execution of a task initiated by its schedule.
- stable releaseA stable release is a software version considered ready for dependable general use rather than experimental use.
- correlationCorrelation is a statistical association describing how two variables vary together.
- CSVCSV is a tabular text format that represents records as rows with fields separated by delimiters.
- data analysisData analysis is the systematic examination, transformation, and interpretation of data to identify patterns or answer questions.
- pandaspandas is a Python library for data manipulation and analysis using labeled tabular structures.
- SciPySciPy is a Python library providing numerical, scientific, and statistical computing functionality.
- structured dataStructured data is information organized according to a defined schema that makes its fields and records machine-readable.
- unstructured textUnstructured text is textual information without a fixed tabular or schema-constrained organization.
- parallelismParallelism is the execution of multiple independent development tasks or plans at the same time.
- /init commandThe /init command is a Claude Code command that generates a CLAUDE.md file from an existing codebase or its documentation.
- live artifactA live artifact is a Claude-generated mini-application that runs in a cloud or local environment and refreshes data when opened.
- prompt patternA prompt pattern is a reusable structure for specifying a desired outcome, its purpose, and applicable constraints.
- command toolThe command tool is a Cowork capability that executes model-generated shell commands inside the local sandbox.
- file write toolThe file write tool is a Cowork capability that writes model-generated content to an approved location on disk.
- versioningVersioning is the practice of preserving and comparing successive states of files or other content.
- plan modePlan mode is an operating mode in which an AI coding agent produces planning documentation instead of implementation code.
These guides are generated from the text of Manning books by Manning's knowledge graph and link to the sections they draw on. The full books are available on liveBook and at manning.com.
