Claude Code in Practice

Class Duration

14 hours of live training delivered over 2-3 days to accommodate your scheduling needs.

Student Prerequisites

Professional software development experience
Comfort with the command line and terminal workflows
Anthropic Claude API access or Claude Code subscription

Target Audience

Senior software engineers, tech leads, and platform engineers who want to master terminal-driven agentic development using Claude Code. Ideal for teams building custom agentic pipelines, automating development workflows, or using Claude Code as the backbone of a CI-integrated AI workflow.

Description

Claude Code is Anthropic's terminal-based agentic coding tool, designed for developers who want a powerful, scriptable AI coding partner that lives in the command line rather than an IDE plugin. This course covers the full Claude Code feature set from setup through advanced agentic workflows: CLAUDE.md project configuration, slash commands, Skills (the unified skill/command model), Plugins for distributable capability bundles, custom tools, hooks for lifecycle automation, subagent orchestration (including background and worktree-isolated subagents), and integration with CI/CD pipelines. Participants build increasingly autonomous workflows throughout the labs, culminating in a complete agentic pipeline for a realistic development task.

Learning Outcomes

Install, authenticate, and configure Claude Code for a professional development environment.
Author effective CLAUDE.md files to give Claude Code persistent project context and behavioral guidelines.
Drive Claude Code with slash commands and interactive session management.
Build and package Skills (the unified .claude/skills/ model) that expose reusable capabilities through a slash-command interface.
Bundle and distribute Plugins to share Skills, hooks, subagents, and MCP server configurations across teams.
Define and invoke custom tools to extend Claude Code's capabilities with internal systems.
Configure hooks for pre/post-task automation and safety guardrails.
Orchestrate subagents for parallelizable and delegated coding tasks, including background and worktree-isolated execution.
Integrate Claude Code into CI pipelines for automated review, test generation, and refactoring.
Apply cost management and observability practices for Claude Code usage.

Training Materials

Comprehensive courseware is distributed online at the start of class. All students receive a downloadable MP4 recording of the training.

Software Requirements

Claude Code CLI installed, Anthropic API key, Git, and the student's preferred language runtime.

Training Topics

Setup and Configuration

Installation and authentication
Model selection and API key management
Shell integration and editor passthrough
Claude Code IDE extensions for VS Code, Visual Studio, JetBrains (IntelliJ, PyCharm, WebStorm, Rider), and Zed
Cost controls and usage monitoring

CLAUDE.md: Project Context Files

Purpose and anatomy of CLAUDE.md
Project conventions, constraints, and workflow rules
Hierarchical CLAUDE.md: repo, directory, and user levels
Version-controlling CLAUDE.md with the team
Coordinating CLAUDE.md with a GitHub Spec Kit constitution for spec-driven workflows

Slash Commands and Interactive Sessions

Built-in slash commands reference
Custom slash commands in .claude/commands/
Session management: resume, export, and clear
Multi-turn task continuation
Interactive vs. non-interactive (headless) modes

Skills

Skills as the unified .claude/skills/ model
Skill anatomy: frontmatter, prompts, allowed tools, model overrides
Building reusable Skills for common engineering tasks
Surfacing Skills as slash commands automatically

Custom Tools

Tool definition syntax and schema
Reading files, running commands, and calling APIs
Tool safety boundaries and confirmation gates
Testing and debugging custom tools

Hooks and Lifecycle Automation

Pre-task and post-task hook types
Input/output transformation hooks
Safety hooks: blocking dangerous commands
Integrating hooks with existing scripts

Subagents and Orchestration

When to spawn a subagent
Subagent scope, isolation, and output collection
Background subagents and worktree isolation
Patterns: fan-out parallelism, pipeline sequencing
Error handling and retry strategies

Plugins: Packaging and Distribution

Plugin anatomy: bundling Skills, hooks, subagents, and MCP servers
Authoring an internal team plugin
Versioning, distribution, and approval workflows
Discovering and installing community plugins from the Claude Skills Marketplace (April 2026)
Skill and plugin governance for team adoption

CI/CD Integration

Running Claude Code in GitHub Actions / GitLab CI
Automated code review on pull requests
Test generation as a CI step
Secrets management in CI context

Observability and Cost Management

Token usage and cost attribution
Logging and tracing agentic sessions
Rate limit management for CI workloads
Responsible usage and policy controls

Workshop

CLAUDE.md authoring lab
Skill + Plugin authoring exercise
Custom tool + hook pipeline exercise
CI integration exercise
Q&A session