Welcome to Claude Code!

You're about to discover the most powerful AI coding assistant available today. Claude Code isn't just another autocomplete tool—it's a full-fledged AI pair programmer that can understand your entire codebase, write complex features, debug issues, and even help with non-coding tasks like data analysis and documentation.

🎉 New in Claude Code 2.0!

Huge updates just released! Check out what's new:

  • 💻 VS Code Extension - Native IDE integration (finally!)
  • 🔄 /rewind - Undo and explore different approaches
  • ⌨️ Keyboard Shortcuts - Tab for thinking, Ctrl+R for history
  • 📎 @-mentions - Quick file references with autocomplete
  • 🎯 Claude Skills - PDF, Excel, and more
  • 🤖 Haiku 4.5 - 5x faster, 1/3 the cost
  • 🌐 Web Search & Fetch - Access current info from the web
  • 💾 Memory Tool - Persistent context across sessions
  • 📏 1M Token Context - Analyze entire codebases (beta)

⚡ Try Claude Code in 5 Minutes

Your first AI pair programming session starts now

1
Open your terminal

Run Claude Code in your project directory

claude
2
Ask Claude to explore

Let Claude understand your codebase

"Help me understand this project"
3
Watch the magic happen

Claude will read your code, analyze structure, and explain it to you in plain English

🎉 You just had your first AI pair programming session!

1

Getting Started

2

Getting Better

3

Advanced

What Makes Claude Code Different?

🧠

Understands Context

Claude Code can read and understand your entire codebase, not just the current file

⚡

Autonomous Actions

Can read, write, edit files, run commands, and search your codebase without constant prompting

🔧

Real Tools

Uses actual file operations, git commands, and shell access—not just text generation

🎯

Task-Oriented

Breaks down complex tasks, creates plans, and executes them step by step

🌐

Extensible

Connect to external services via MCP, customize with hooks, and create slash commands

💼

Beyond Coding

Data analysis, report generation, documentation, and more

Your Learning Path

Phase 1: Getting Started (30 minutes)

Learn the basics: how to communicate with Claude Code, your first tasks, understanding the interface, and basic workflows.

Click to start →

Phase 2: Getting Better (1-2 hours)

Deep dive into tools, file operations, codebase understanding, and common development workflows.

Click to continue →

Phase 3: Advanced (Ongoing)

MCPs, hooks, custom commands, automation, and advanced integration patterns.

Click to explore →

✅ Your First Day with Claude Code

Complete these tasks to build a solid foundation. Your progress is automatically saved!

0/5 completed
Pro Tip: The best way to learn Claude Code is by using it! Start with simple tasks and gradually take on more complex challenges. Click "Getting Started" above to begin your journey.

📺 Watch & Learn

Official Learning Resources:

👋 Welcome to Your Claude Code Journey!

You've learned what makes Claude Code special and seen your learning path ahead.

Ready to write your first commands?

Getting Started

Learn the basics in 30 minutes

What's New in Claude Code 2.0

🎉 Claude Code 2.0 is Here!

Released September 29, 2025 - The biggest update yet with powerful new features

Claude Code 2.0 brings a completely redesigned experience with native IDE integration, powerful workflow commands, enhanced keyboard shortcuts, and intelligent new features that make you even more productive.

🚀 Major New Features

💻 Native VS Code Extension

The #1 requested feature is here!

Claude Code now runs natively inside Visual Studio Code, bringing AI pair programming directly into your favorite editor.

Key Benefits:

  • Seamless Integration: No more switching between terminal and editor
  • Visual Diffs: See changes side-by-side with syntax highlighting
  • Context Awareness: Claude knows your current file and selection
  • One-Click Actions: Accept or reject changes with a single click
  • Conversation History: Access past sessions right in VS Code

Getting Started with VS Code Extension

# Install the extension 1. Open VS Code 2. Go to Extensions (Cmd+Shift+X / Ctrl+Shift+X) 3. Search for "Claude Code" 4. Click Install # Or install via command line code --install-extension anthropic.claude-code # Start using Claude Code in VS Code # Look for the Claude icon in your sidebar!

⚡ Powerful Workflow Commands

New commands that supercharge your development workflow:

🔄 /rewind

Undo code changes and go back in conversation history

/rewind # Step back through changes # Perfect for trying different approaches!

📊 /usage

Track your usage and plan limits in real-time

/usage # See tokens used, cost, limits # Monitor your Claude Max/Pro plan

📤 /export

Export conversations to share with your team

/export # Export full conversation # Share learnings and solutions

📋 /statusline

Customize your prompt with custom status information

/statusline # Add git branch, time, etc # Personalize your experience

🔍 /context

Debug context issues and see what Claude knows

/context # See token usage breakdown # Optimize your prompts

📝 /memory

Manage persistent memory across sessions

/memory # Edit memory files directly # Control what Claude remembers

⌨️ Keyboard Shortcuts

Master these shortcuts to work faster than ever:

Toggle Thinking Mode Tab

Enable extended reasoning (sticky across sessions)

History Search Ctrl+R

Search previous commands like bash/zsh

Background Command Ctrl+B

Run bash commands in background (dev servers, logs)

Toggle Transcript Ctrl+O

View full conversation transcript

Edit in $EDITOR Ctrl+G

Open prompt in your system text editor

Mode Switching (Windows) Shift+Tab

Switch modes on Windows native install

Pro Tip: Press Tab once to enable thinking for complex problems, then press it again to disable. The setting persists across sessions!

📎 @-mentions for Quick References

Quickly reference files and directories in your prompts with @-mentions:

Using @-mentions

# Mention a specific file "Review @src/components/Button.tsx for accessibility issues" # Mention multiple files "Compare @api/users.ts and @api/auth.ts" # Mention directories "Analyze all components in @src/components/" # Mention images (for analysis) "What's in this screenshot? @screenshots/error.png" # Tab completion works with @-mentions! "Fix the bug in @sr" [press Tab to autocomplete]
Why it matters: @-mentions provide Claude with immediate file context without waiting for file reads. Tab completion makes it fast and accurate!

🎯 Claude Skills

Extend Claude Code with specialized skills for specific tasks:

📄 PDF Skill

Extract text, create PDFs, merge documents, handle forms

📊 Excel Skill

Read, write, and manipulate spreadsheet data

🎨 Canvas Design

Create visual designs and artwork programmatically

🔧 Custom Skills

Build your own skills for specialized workflows

Using Skills

# Claude automatically activates skills when needed "Extract all tables from this PDF and convert to CSV" # → PDF skill activates automatically "Create a poster for our hackathon event" # → Canvas design skill activates # Or explicitly invoke a skill /skill pdf

🤖 Enhanced Models & Subagents

Haiku 4.5 - Fast & Cost-Effective

The newest addition to the Claude family is now available in Claude Code:

  • 4-5x faster than Sonnet 4.5
  • 73.3% on SWE-bench Verified - Sonnet 4-level coding at 1/3 the cost
  • Perfect for: High-volume tasks, real-time applications, cost-sensitive workloads
  • 200K context window (1M on Developer Platform)

SonnetPlan Mode

Get the best of both worlds with automatic model selection:

  • Uses Sonnet 4.5 for planning (complex reasoning)
  • Uses Haiku 4.5 for execution (fast, cost-effective implementation)
  • Automatically enabled when you select Haiku 4.5
# Select Haiku 4.5 to enable SonnetPlan mode /model # Choose "Haiku 4.5" # Claude will use Sonnet for thinking, Haiku for doing!

Enhanced Subagents

  • Plan Subagent: Dedicated agent for creating detailed implementation plans
  • Explore Subagent (Haiku-powered): Efficiently searches your codebase to save context
  • Resumable Subagents: Claude can now resume and continue previous subagent sessions
  • Dynamic Model Selection: Claude chooses the right model for each subagent task

❓ Interactive Question Tool

Claude can now ask you questions to clarify requirements and gather information:

How It Works

You: "Add authentication to the app"

Claude: "Which authentication method would you prefer?"

  • OAuth 2.0 with social providers (Google, GitHub)
  • JWT with email/password
  • Magic link authentication
  • Other (specify)

You select an option, and Claude proceeds with clarity!

Why it matters: No more guessing! Claude asks clarifying questions to ensure it builds exactly what you need, especially in Plan mode.

🌐 Web Search & Web Fetch

Released: May 7 & September 10, 2025

Claude Code now has built-in web access for current information:

Web Search

  • Access current information - Get real-time data, news, and updates
  • Research capabilities - Search the web for technical documentation, articles, and more
  • Domain filtering - Include or block specific websites in search results

Web Fetch

  • Retrieve full page content - Fetch complete web pages and PDFs from URLs
  • Process web documents - Extract and analyze content from online sources

Using Web Features

# Claude automatically uses web search when needed "What are the latest Python 3.13 features released in 2025?" # Fetch and analyze web content "Analyze the API documentation at https://docs.example.com/api" # Research best practices "Find the latest React Server Components best practices"

💾 Memory Tool - Persistent Context

Released: September 29, 2025

Claude can now remember information across conversations and sessions:

  • Persistent memory - Store preferences, project context, and team conventions
  • Cross-session recall - Information persists even when you restart Claude Code
  • Manage with /memory - Edit and review what Claude remembers
  • Automatic learning - Claude learns from your corrections and preferences

How Memory Works

# Claude automatically stores important context "Remember: we use TypeScript strict mode and prefer functional components" # Later, in a new session... "Create a new component" # Claude automatically uses TypeScript and functional style! # Manage your memory /memory # View and edit what Claude remembers
Why it matters: Stop repeating yourself! Claude remembers your preferences, coding style, and project conventions automatically.

🧠 Advanced Context Management

Context Editing (September 29, 2025)

Claude now automatically manages conversation context for optimal performance:

  • Automatic optimization - Claude removes unnecessary thinking blocks and old context
  • Smarter context window use - Keep conversations going longer without hitting limits
  • Seamless experience - Happens automatically in the background

Files API (May 22, 2025)

Upload and work with files more efficiently:

  • Direct file uploads - Share files for analysis, processing, or reference
  • Code execution context - Files available during code execution
  • Multi-file operations - Work with multiple documents simultaneously

Interleaved Thinking (May 22, 2025)

Claude can now think between tool calls for better decision-making:

  • Reasoning during execution - Think and adjust strategy as work progresses
  • Better tool selection - Choose the right tools based on intermediate results
  • Adaptive workflows - Adjust approach based on what's discovered

📏 1M Token Context Window (Beta)

Released: August 12, 2025

Sonnet 4.5 now supports up to 1 million tokens of context in beta:

  • Massive codebase support - Analyze entire repositories at once
  • Long conversation threads - Work on complex projects without losing context
  • Extended documentation - Include full API docs and specifications
  • Beta access - Enable via API header on supported platforms
Pro Tip: The 1M token context window is currently in beta. Use it for analyzing large codebases or working with extensive documentation. Standard 200K context is still recommended for most use cases.

🔄 Upgrading from 1.0

Already using Claude Code? Here's what you need to know:

  • ✅ Automatic update: Run npm install -g @anthropic-ai/claude-code@latest
  • ✅ No breaking changes: All your existing workflows, hooks, and MCPs work as before
  • ✅ VS Code extension: Install separately for IDE experience
  • ✅ Settings preserved: Your configuration carries over automatically
# Update Claude Code to 2.0 npm install -g @anthropic-ai/claude-code@latest # Verify you're on 2.0 claude --version # Should show 2.0.x or higher

📚 Learn More

VS Code Extension

Get started with the native VS Code experience

Read the guide

2.0 Release Notes

Full details on all changes and improvements

View changelog

Keyboard Shortcuts

Master all the new productivity shortcuts

See all shortcuts

Claude Skills

Learn about Skills and build your own

Explore Skills

🎉 You're Up to Date!

You now know about all the powerful new features in Claude Code 2.0.

Ready to install and get started?

Installation Guide

Set up Claude Code in minutes

How to Install Claude Code

Choose how you want to use Claude Code. The VS Code extension is the easiest way to get started!

🆕 VS Code Extension (Recommended)

Why Choose VS Code Extension?

  • ✨ Easiest to install - Works in seconds
  • 💻 Native IDE integration - No terminal switching
  • 👁️ Visual diffs - See changes side-by-side
  • ⚡ One-click actions - Accept/reject with a button
  • 📜 Conversation history - Built into VS Code

Step 1: Prerequisites

  • Visual Studio Code installed (Download here)
  • An Anthropic API key, Claude Pro, or Claude Max subscription

Step 2: Install the Extension

Option A: Install from VS Code

  1. Open VS Code
  2. Click the Extensions icon (or press Cmd+Shift+X / Ctrl+Shift+X)
  3. Search for "Claude Code"
  4. Click Install on the official Anthropic extension

Option B: Install via Command Line

code --install-extension anthropic.claude-code

Step 3: Get Your API Key (If Needed)

If you're using Anthropic API:

  • Visit console.anthropic.com
  • Sign in or create an account
  • Navigate to "API Keys" in settings
  • Click "Create Key" and copy it

If you have Claude Pro or Claude Max, you can sign in directly!

Step 4: Configure Authentication

When you first open the extension:

  1. Look for the Claude icon in your VS Code sidebar
  2. Click it to open Claude Code
  3. Follow the prompts to sign in or enter your API key
  4. Choose your preferred model (Sonnet 4.5 recommended)

Step 5: Start Using Claude Code!

You're all set! Try your first prompt:
  • "Help me understand this project"
  • "Add error handling to this function"
  • "Refactor this code to be more readable"

Quick Tips:

  • Right-click any file to ask Claude about it
  • Select code and ask Claude to explain or improve it
  • Use Cmd+K / Ctrl+K for quick inline edits
Want the CLI too? You can use both the VS Code extension AND the terminal version. Just follow the "CLI via Anthropic API" tab to install the command-line version.

Direct Install via Anthropic API

Step 1: Check Prerequisites

Ensure you have the following installed:

  • Node.js version 18.0.0 or higher
  • npm or yarn package manager
  • An active Anthropic API account

Check your Node.js version:

node --version

Step 2: Get Your API Key

  • Visit console.anthropic.com
  • Sign in or create an account
  • Navigate to "API Keys" in the settings
  • Click "Create Key" and copy your new API key
  • Store it securely - you'll need it in the next step

Step 3: Install Claude Code

Install globally using npm:

npm install -g @anthropic-ai/claude-code

Or using yarn:

yarn global add @anthropic-ai/claude-code

Step 4: Configure Your API Key

Set your API key as an environment variable:

On macOS/Linux:

export ANTHROPIC_API_KEY='your-api-key-here'

On Windows (PowerShell):

$env:ANTHROPIC_API_KEY='your-api-key-here'

Make it permanent (add to ~/.zshrc or ~/.bashrc):

echo 'export ANTHROPIC_API_KEY="your-api-key-here"' >> ~/.zshrc

Step 5: Verify Installation

Run Claude Code to verify everything is set up correctly:

claude --version

Start your first session:

claude

✓ You're ready to go! Claude Code should now be running.

Install via Google Vertex AI

Step 1: Check Prerequisites

  • Node.js version 18.0.0 or higher
  • npm or yarn package manager
  • A Google Cloud Platform account with billing enabled
  • gcloud CLI installed and configured

Step 2: Set Up Google Cloud Project

Create or select a GCP project:

gcloud projects create YOUR_PROJECT_ID

Set the project as default:

gcloud config set project YOUR_PROJECT_ID

Enable required APIs:

gcloud services enable aiplatform.googleapis.com

Step 3: Enable Vertex AI and Claude Access

  • Visit the Vertex AI console
  • Navigate to Model Garden
  • Search for "Claude" models
  • Request access to Claude models (if needed)
  • Wait for approval (typically within 24 hours)

Step 4: Set Up Authentication

Create a service account:

gcloud iam service-accounts create claude-code-sa \ --display-name="Claude Code Service Account"

Grant necessary permissions:

gcloud projects add-iam-policy-binding YOUR_PROJECT_ID \ --member="serviceAccount:claude-code-sa@YOUR_PROJECT_ID.iam.gserviceaccount.com" \ --role="roles/aiplatform.user"

Create and download credentials:

gcloud iam service-accounts keys create ~/claude-key.json \ --iam-account=claude-code-sa@YOUR_PROJECT_ID.iam.gserviceaccount.com

Step 5: Install Claude Code

npm install -g @anthropic-ai/claude-code

Step 6: Configure Claude Code for Vertex AI

Set environment variables:

export GOOGLE_APPLICATION_CREDENTIALS="$HOME/claude-key.json" export ANTHROPIC_VERTEX_PROJECT_ID="YOUR_PROJECT_ID" export ANTHROPIC_VERTEX_REGION="us-central1"

Start Claude Code with Vertex AI:

claude --provider vertex

✓ Claude Code is now running via Google Vertex AI!

Step 7: Enable Telemetry (Optional but Recommended)

Enable telemetry to help Anthropic improve Claude Code and provide better support. This is particularly useful for enterprise deployments to track usage and performance.

Set telemetry environment variables:

export CLAUDE_TELEMETRY_ENABLED=true export CLAUDE_TELEMETRY_ENDPOINT="your-org-telemetry-endpoint"

Configure telemetry in your .claude.json file:

{ "telemetry": { "enabled": true, "endpoint": "your-org-telemetry-endpoint", "includeUsageMetrics": true, "includePerformanceMetrics": true } }

Verify telemetry is enabled:

claude --show-config

Note:

Telemetry data is anonymized and helps improve Claude Code's performance, especially for Vertex AI deployments. No code content is transmitted, only usage patterns and performance metrics.

Additional Resources

Install via AWS Bedrock

Step 1: Check Prerequisites

  • Node.js version 18.0.0 or higher
  • npm or yarn package manager
  • An AWS account with Bedrock access
  • AWS CLI installed and configured

Step 2: Enable AWS Bedrock

Sign in to the AWS Console and:

  • Navigate to AWS Bedrock service
  • Select a supported region (us-east-1, us-west-2, etc.)
  • Go to "Model access" in the left sidebar
  • Request access to Claude models (Sonnet, Opus, Haiku)
  • Wait for approval (typically instant or within minutes)

Step 3: Configure AWS Credentials

Set up AWS CLI credentials (if not already configured):

aws configure

You'll be prompted to enter:

  • AWS Access Key ID
  • AWS Secret Access Key
  • Default region (e.g., us-east-1)
  • Output format (json recommended)

Step 4: Set Up IAM Permissions

Ensure your IAM user/role has these permissions:

{ "Version": "2012-10-17", "Statement": [ { "Effect": "Allow", "Action": [ "bedrock:InvokeModel", "bedrock:InvokeModelWithResponseStream" ], "Resource": "arn:aws:bedrock:*::foundation-model/anthropic.claude*" } ] }

Step 5: Install Claude Code

npm install -g @anthropic-ai/claude-code

Step 6: Configure Claude Code for Bedrock

Set your AWS region (if different from default):

export AWS_REGION="us-east-1"

Start Claude Code with Bedrock:

claude --provider bedrock

Or specify region directly:

claude --provider bedrock --region us-west-2

✓ Claude Code is now running via AWS Bedrock!

Additional Resources

Quick Tips

  • Need help? Run claude --help for all available commands
  • Update Claude Code: Run npm update -g @anthropic-ai/claude-code
  • Switch providers: Use the --provider flag when starting Claude
  • Troubleshooting? Check the Troubleshooting tab for common issues

Phase 1: Getting Started

Your First Conversation

Claude Code works through natural conversation. Simply type your request and Claude will break it down, create a plan, and execute it.

Example: Your First Task

You: "Can you help me understand what this project does?"

# Claude will: 1. Read your README.md 2. Scan the project structure 3. Identify key files and directories 4. Provide a summary of the project's purpose and architecture

Example: Making Changes

You: "Add error handling to the login function"

# Claude will: 1. Search for the login function 2. Read the file containing it 3. Analyze the current implementation 4. Add appropriate try-catch blocks 5. Update the code with proper error messages

⚠️ Common Mistake: Being Too Vague

❌ Less Effective: "Fix the code"

✅ More Effective: "Add input validation to the email field in the registration form"

Why: Specific requests help Claude understand exactly what you need and provide better results faster.

Understanding the Interface

How Claude Communicates

  • Text responses: Claude explains what it's doing and shares findings
  • Tool usage: You'll see when Claude reads files, edits code, or runs commands
  • Code blocks: Changes are shown clearly with syntax highlighting
  • Thinking process: Claude shows its reasoning as it works
Tip: Claude Code operates in your terminal. You can run it with claude command, and you'll have a conversation-based interface where you can ask questions, request changes, and get work done.

Basic Commands & Workflows

Essential Commands

# Start Claude Code claude # Get help /help # Clear conversation history /clear # Exit Claude Code /exit or Ctrl+C

Productivity Shortcuts (New in 2.0!)

# Press Tab to toggle thinking mode (sticky setting!) # Great for complex problems that need deep reasoning # Press Ctrl+R to search command history # Just like in bash/zsh - super handy! # Press Ctrl+B to run commands in background # Perfect for dev servers: "npm run dev" in background

💡 Pro Tip: Learn these shortcuts early - they'll save you tons of time!

@-mentions for Quick File References

Speed Up Your Workflow

# Mention files directly in your prompts "Review @src/app.js for security issues" # Use Tab completion with @-mentions "Fix the bug in @sr" [press Tab → auto-completes to @src/] # Mention multiple files "Compare @api/v1/users.ts and @api/v2/users.ts"

Why it's awesome: Gives Claude immediate file context without waiting for searches!

Your First Real Tasks

Task 1: Explore a Codebase

"Help me understand this codebase. Show me the overall structure and main components." # Claude will explore the project, read key files, and provide a comprehensive overview

Task 2: Fix a Bug

"The user authentication is failing. Can you debug and fix it?" # Claude will: # 1. Find authentication-related code # 2. Analyze the implementation # 3. Identify the issue # 4. Fix the bug # 5. Explain what was wrong

Task 3: Add a Feature

"Add a dark mode toggle to the settings page" # Claude will: # 1. Create a task list # 2. Add the toggle component # 3. Implement state management # 4. Update styles # 5. Test the implementation

Task 4: Refactor Code

"Refactor the data processing module to use async/await instead of callbacks" # Claude will systematically update the code while maintaining functionality

💡 Pro Tip: Give Claude Time to Understand

Don't immediately jump into making changes. Let Claude read and analyze your codebase first.

✅ Try this: "Help me understand this project before we make changes"

This helps Claude provide better context-aware suggestions and avoid misunderstandings.

How Claude Code Works: The Tools

Claude has access to several powerful tools that let it interact with your codebase:

📖 Read

Read any file in your project. Claude can read code, configs, docs, even images and PDFs.

✏️ Edit

Make precise changes to files. Claude shows you exactly what's being changed.

📝 Write

Create new files. Usually Claude prefers to edit existing files when possible.

🔍 Glob

Find files by pattern. Quick pattern matching for locating files.

🔎 Grep

Search file contents. Find code patterns, function definitions, specific text.

⚙️ Bash

Run terminal commands. Git, npm, build tools, tests—anything you can run in terminal.

Pro Tip: You don't need to tell Claude which tools to use. Just describe what you want, and Claude will choose the right tools automatically!

⚠️ Common Mistake: Not Reviewing Changes

Claude shows you all changes before they're applied. Always review them!

✅ Best practice: Read the changes carefully, ask questions if anything is unclear, then confirm.

You can always undo changes with git if something doesn't look right: git restore <file>

Best Practices for Beginners

✅ Be Clear and Specific

Good: "Add input validation to the email field in the registration form"

Less good: "Fix the form"

✅ Let Claude Explore First

Claude can read and understand your codebase. Let it explore before making changes.

✅ Review Changes

Claude shows you all changes. Review them before committing.

✅ Ask Questions

"Why did you make this change?" or "What does this function do?" - Claude loves explaining!

✅ Iterate and Refine

If something isn't quite right, just say so. "Can you make this more efficient?" or "Add comments to explain this logic"

💡 Pro Tip: Don't Expect Instant Perfection

Claude is powerful but may need iteration. This is normal and expected!

✅ Try this:

• "Can you make this more efficient?"

• "Add comments to explain this logic"

• "Let's try a different approach"

Iteration leads to better results than trying to get everything perfect in one shot.

⚠️ Common Mistake: Information Overload

Don't try to give Claude your entire codebase context in one message.

✅ Better: Let Claude explore naturally through the conversation, then provide specific context when needed.

Claude can read files and search your codebase—trust it to find what it needs!

Checkpoint: You now know the basics! Try running a few simple tasks in your own projects. When you're comfortable, move on to "Getting Better" to unlock more power.

🎉 You've Completed Getting Started!

You learned: How to communicate with Claude, basic commands, available tools, and best practices for beginners.

Ready to level up?

Getting Better

Master file operations, git workflows, and development patterns

Phase 2: Getting Better

Mastering File Operations

Reading Files Strategically

Understanding Project Structure

You: "Show me all the React components in this project"

# Claude will use Glob to find all .jsx/.tsx files # Then provide a categorized overview

Finding Specific Code

You: "Where is the database connection configured?"

# Claude will Grep for keywords like: # - createConnection, mongoose.connect, pool.query # - DATABASE_URL, connection strings # Then read the relevant files

Making Complex Edits

Refactoring Across Multiple Files

You: "Rename the 'getUserData' function to 'fetchUserProfile' everywhere"

# Claude will: 1. Search for all occurrences 2. Identify files that need changes 3. Make consistent edits across all files 4. Update imports and references

Working with Git

Creating Commits

You: "Commit these changes with a good message"

# Claude will: git status # Check what's changed git diff # Review the changes git add # Stage relevant files git commit -m "Add user authentication with JWT"
Note: Claude writes descriptive commit messages following best practices!

Creating Pull Requests

You: "Create a pull request for this feature"

# Claude will: 1. Review all commits in the branch 2. Write a comprehensive PR description 3. Create test plan 4. Push to remote 5. Use gh CLI to create the PR

Powerful Workflow Commands (2.0 Features)

Master these game-changing commands introduced in Claude Code 2.0:

🔄 /rewind

Undo and explore alternatives

/rewind

Perfect when you want to try a different approach or step back through changes. Like git reset but for your conversation!

📊 /usage

Monitor your token usage

/usage

Track tokens consumed, estimated costs, and your plan limits in real-time. Great for staying within budget!

📤 /export

Share conversations

/export

Export full conversation history to share solutions with your team or document your work.

🔍 /context

Debug context usage

/context

See exactly what's in Claude's context window and optimize your token usage.

📝 /memory

Manage persistent memory

/memory

Edit what Claude remembers across sessions. Perfect for project-specific conventions!

📋 /statusline

Customize your prompt

/statusline

Add git branch, timestamp, or custom info to your terminal prompt.

Real-World Workflow Example

# 1. Start a feature "Add user profile page with avatar upload" # 2. Claude implements it # 3. You realize you want a different approach /rewind # 4. Try alternative approach "Actually, let's use a modal instead of a separate page" # 5. Check your usage /usage # Tokens: 45K/200K (23% of context used) # 6. When done, export to share with team /export

💡 Pro Workflow Tips

  • /rewind + experiment: Don't be afraid to try different approaches - you can always rewind!
  • /context regularly: Check context usage periodically to avoid hitting limits
  • /export after breakthroughs: Document successful solutions for future reference
  • /memory for conventions: Store coding standards, naming patterns, or project-specific rules

Development Workflows

🔄 Test-Driven Development

"Write tests for the authentication module, then implement the functionality" # Claude will: # 1. Create comprehensive test cases # 2. Implement the feature to pass tests # 3. Run tests to verify # 4. Refine as needed

🐛 Debugging Workflow

"The app crashes when I click the submit button. Can you debug?" # Claude's process: # 1. Find the submit button handler # 2. Trace the execution path # 3. Look for error conditions # 4. Check logs/console output if available # 5. Identify and fix the issue # 6. Add defensive code to prevent future issues

📦 Dependency Management

"Update all dependencies and fix any breaking changes" # Claude will: npm outdated # Check what's outdated npm update # Update packages npm run test # Test for breakage # Fix any issues that arise # Update code for breaking changes

🏗️ Code Architecture

"Analyze the architecture and suggest improvements" # Claude will: # 1. Map out the current architecture # 2. Identify patterns and anti-patterns # 3. Suggest concrete improvements # 4. Help implement refactoring if requested

Advanced Communication Patterns

Giving Context

Better: Provide Context

"We're using Redux for state management. Add a new action to handle user logout, making sure to clear the user data from the store and redirect to the login page."

Breaking Down Complex Tasks

Complex Feature Request

"Build a user dashboard with: - User profile display - Recent activity feed - Settings panel - Analytics charts Please create a todo list first, then we'll work through each component." # Claude will create a structured plan and track progress

Iterative Refinement

Refining Output

Initial: "Create a login form"

Refine: "Add form validation"

Refine: "Add a 'Remember Me' checkbox"

Refine: "Add 'Forgot Password' link"

Refine: "Make it mobile responsive"

Advanced Prompting Techniques

Unlock more powerful responses from Claude by using these specialized prompting techniques.

Extended Thinking (Deep Analysis)

⚡ Quick Tip: Press Tab to Enable Thinking!

The easiest way to activate extended thinking is to press Tab in the input. The setting is sticky—it stays enabled until you press Tab again!

# Just press Tab before or after typing your prompt "Design a caching strategy for our API" [Tab pressed] # Claude will now think deeply about the problem! # Press Tab again to disable thinking mode

When you need Claude to think deeply about complex problems, you can request extended reasoning. Claude will break down the problem step-by-step before providing a solution.

When to use thinking mode:

  • Complex architectural decisions
  • Debugging difficult issues
  • Algorithm design
  • Comprehensive code reviews
  • Performance optimization analysis

Example: Request Deep Thinking via Prompt

"Think deeply about the best approach to refactor our authentication system. Consider security, performance, maintainability, and migration complexity." # Or use these prompting patterns: "Let's think through this step-by-step..." "Analyze this problem carefully and show your reasoning..." "Take your time to consider all angles..." "ultrathink about the best solution..."

Example: Architecture Decision with Reasoning

"We need to choose between microservices and monolith for our new project. Think through the tradeoffs carefully: - Team size: 8 developers - Expected scale: 100k users in year 1 - Deployment frequency: Multiple times per week - Current expertise: Strong in Node.js, learning containers Show your reasoning process and recommend an approach." # Claude will provide structured reasoning before the recommendation

Step-by-Step Problem Solving

Requesting Methodical Approach

"Walk me through debugging this memory leak step-by-step. Start by identifying potential causes, then guide me through testing each hypothesis." # Claude will break down the investigation into clear steps

Comparative Analysis

Example: Comparing Approaches

"Compare these three state management approaches for our React app: 1. Redux with Redux Toolkit 2. Zustand 3. React Context + useReducer Analyze based on: learning curve, boilerplate, performance, DevTools support, and team size. Show your analysis in a structured comparison." # Claude will provide detailed comparison tables and reasoning

Plan Mode for Large Tasks

For very large or complex tasks, Claude Code has a "plan mode" that creates a detailed plan before executing. This helps with:

  • Multi-file refactoring
  • Large feature implementations
  • System-wide changes
  • Migration projects

How to use: Simply ask Claude to create a plan first, or ask about complex tasks that benefit from planning.

Example: Request Planning

"I want to migrate our entire app from JavaScript to TypeScript. Create a detailed plan first, breaking this into phases." # Or: "Let's plan out this refactoring before we start..." "Create a todo list for this feature implementation..."

Prompting Best Practices

🎯 Be Specific About Reasoning

Ask Claude to "show your reasoning" or "explain your thinking" for transparency

📋 Request Structure

Ask for "step-by-step" or "structured analysis" for organized output

🔍 Request Analysis First

For complex tasks: "Analyze first, then implement" gets better results

⚖️ Ask for Tradeoffs

"What are the pros and cons?" or "What are the tradeoffs?" for balanced decisions

🎨 Specify Constraints

Mention team size, expertise, timeline, or other constraints for realistic solutions

🔄 Encourage Iteration

"Let's refine this..." or "What if we tried..." for collaborative problem-solving

Pro Tip: For really complex problems, combine techniques: "Think deeply about the architecture (extended thinking), create a plan (plan mode), then we'll implement step-by-step (iterative)." This layered approach yields the best results for challenging tasks.

Understanding Claude's Capabilities

📚 Multi-file Understanding

Claude can understand relationships between files, trace imports, and see how components interact.

🔄 Context Awareness

Claude remembers the conversation. Reference previous changes or discussions naturally.

🎯 Pattern Recognition

Claude learns your coding patterns and can apply them consistently across your project.

🧪 Testing

Claude can write, run, and debug tests for your code automatically.

📖 Documentation

Generate docs, add comments, create README files—Claude understands documentation best practices.

🔍 Code Review

Ask Claude to review code for bugs, performance issues, security concerns, or best practices.

Real-World Workflow Examples

Morning Workflow: Catch Up on Changes

"Show me what changed in the codebase since yesterday" git log --since="yesterday" --oneline git diff HEAD@{yesterday} # Claude will summarize key changes and their impact

Feature Development Workflow

# 1. Planning Phase "I need to add a shopping cart feature. Let's plan this out." # 2. Implementation Phase "Let's start with the cart state management" "Now add the cart UI component" "Connect it to the backend API" # 3. Testing Phase "Write tests for the cart functionality" "Run the tests and fix any failures" # 4. Refinement Phase "Add loading states and error handling" "Make sure it's responsive on mobile" # 5. Documentation Phase "Document the cart API and add code comments" # 6. Ship It! "Create a PR for the shopping cart feature"

Bug Fix Workflow

"Users are reporting that images aren't loading. Let's investigate." # Claude will: # 1. Find image loading code # 2. Check for common issues (paths, permissions, async problems) # 3. Look at recent changes that might have broken it # 4. Fix the issue # 5. Add defensive code to prevent recurrence # 6. Suggest adding tests
Level Up! You're now comfortable with Claude Code's core capabilities. Ready to explore advanced features? Head to the "Advanced" tab!

📺 Advanced Learning

Deep Dive Resources:

🚀 You're Now at Intermediate Level!

You learned: File operations, git workflows, development patterns, and communication techniques.

Time to unlock advanced superpowers!

Advanced

MCPs, hooks, custom commands, and automation

Phase 3: Advanced

Model Context Protocol (MCP)

MCPs allow Claude Code to connect to external services and data sources, dramatically expanding its capabilities.

What is MCP?

MCP is a protocol that lets Claude Code interact with external systems like databases, APIs, cloud services, and more. Think of it as giving Claude superpowers to access data and services beyond your local filesystem.

Popular MCP Servers

🗄️ Database MCPs

PostgreSQL, MySQL, MongoDB, SQLite

Use case: Query databases, analyze data, run migrations

☁️ Cloud Provider MCPs

AWS, Google Cloud, Azure

Use case: Manage cloud resources, deploy applications, monitor services

📊 Data Source MCPs

Google Drive, Notion, Airtable

Use case: Access documents, sync data, generate reports

🔧 Development Tool MCPs

GitHub, GitLab, Jira, Linear

Use case: Manage issues, review PRs, track projects

🌐 API MCPs

REST APIs, GraphQL endpoints

Use case: Integrate with third-party services

📈 Analytics MCPs

Google Analytics, Mixpanel, custom metrics

Use case: Analyze usage, generate insights, create dashboards

Setting Up an MCP Server

Example: PostgreSQL MCP

# 1. Create or edit ~/.mcp.json { "mcpServers": { "postgres": { "command": "npx", "args": [ "-y", "@anthropic-ai/mcp-server-postgres", "postgresql://user:pass@localhost/mydb" ] } } } # 2. Use it in Claude Code "Query the users table and show me the most recent signups" # Claude will now use the MCP to query your database!

Example: Google Drive MCP

# Configuration in ~/.mcp.json { "mcpServers": { "gdrive": { "command": "npx", "args": ["-y", "@anthropic-ai/mcp-server-gdrive"] } } } # Usage "Find all spreadsheets in my Google Drive related to Q4 revenue" "Create a summary report from the data in those spreadsheets"
Pro Tip: Explore available MCP servers at github.com/modelcontextprotocol/servers

🎯 Real-World MCP Use Cases

  • Data Analysis: "Compare sales data from our PostgreSQL database with the projections in Google Sheets"
  • DevOps: "Check the status of all our AWS EC2 instances and alert me if any are overutilized"
  • Project Management: "Show me all high-priority bugs from Jira and create a summary report"
  • Documentation: "Pull all our API docs from Notion and generate a unified developer guide"

Hooks: Customizing Claude's Behavior

Hooks let you run shell commands in response to events, allowing you to customize Claude Code's workflow.

Types of Hooks

🎯 user-prompt-submit-hook

Runs after you submit a message. Perfect for:

  • Running linters before Claude edits code
  • Checking git status
  • Validating environment setup

⚡ tool-call-hook

Runs before specific tool calls. Use to:

  • Backup files before edits
  • Run formatters after writes
  • Validate changes

Setting Up Hooks

Example: Run Prettier After File Edits

# In your .claude/config.json { "hooks": { "tool-call-hook": { "Edit": "prettier --write {{file_path}}" } } } # Now, whenever Claude edits a file, Prettier auto-formats it!

Example: Git Status Before Each Prompt

# Show git status to give Claude context about changes { "hooks": { "user-prompt-submit-hook": "git status --short" } }

Example: Run Tests Before Commits

# Prevent commits if tests fail { "hooks": { "tool-call-hook": { "Bash": { "pattern": "git commit", "command": "npm test" } } } }
Pro Tip: Hooks can return feedback that Claude will see and react to. If a test fails, Claude will see the error and can fix it!

Custom Slash Commands

Create custom commands that expand into detailed prompts, perfect for repeated workflows.

Creating Slash Commands

Example: Code Review Command

# Create .claude/commands/review.md Please perform a comprehensive code review of the current changes: 1. Check for bugs and logic errors 2. Review for security vulnerabilities 3. Assess performance implications 4. Verify code follows our style guide 5. Check for proper error handling 6. Ensure adequate test coverage 7. Review documentation and comments Provide a structured review with specific file:line references. # Usage: /review

Example: Documentation Generator

# Create .claude/commands/document.md Generate comprehensive documentation for this project: 1. Create/update README.md with: - Project overview - Installation instructions - Usage examples - API documentation 2. Add JSDoc comments to all functions 3. Create CONTRIBUTING.md with guidelines 4. Document environment variables needed # Usage: /document

Example: Performance Audit

# Create .claude/commands/perf-audit.md Conduct a performance audit: 1. Identify slow database queries 2. Check for N+1 query problems 3. Review bundle size and imports 4. Look for unnecessary re-renders (React) 5. Check for memory leaks 6. Analyze algorithm complexity 7. Suggest specific optimizations # Usage: /perf-audit

Powerful Slash Command Ideas

/onboard

Generate comprehensive onboarding docs for new team members

/security-check

Audit code for security vulnerabilities

/deploy-prep

Pre-deployment checklist and verification

/api-test

Generate comprehensive API test suite

/refactor-plan

Create detailed refactoring plan for a module

/dependency-audit

Check for outdated/vulnerable dependencies

Plugins

Plugins allow you to extend Claude Code with custom functionality that can be shared across projects and teams. They can add custom commands, agents, hooks, and MCP servers all in one package.

What Are Plugins?

Think of plugins as pre-packaged extensions that bundle together:

  • Custom Commands: Slash commands for specific workflows
  • Agents: Specialized AI agents for particular tasks
  • Hooks: Automated event handlers
  • MCP Servers: Connections to external tools and services

Plugins make it easy to share best practices and tooling across your entire team or organization.

Plugin Structure

Basic Plugin Layout

my-plugin/ ├── .claude-plugin/ │ └── plugin.json # Plugin metadata ├── commands/ # Custom slash commands │ ├── review.md │ └── deploy.md ├── agents/ # Specialized agents │ └── code-reviewer.md ├── hooks/ # Event handlers │ └── pre-commit.sh └── mcp/ # MCP server configs └── database.json

Plugin Manifest (plugin.json)

{ "name": "my-awesome-plugin", "version": "1.0.0", "description": "Adds code review and deployment workflows", "author": "Your Name", "homepage": "https://github.com/yourname/plugin" }

Installing Plugins

Method 1: Interactive Menu (Recommended)

# Browse available plugins /plugin marketplace # Install a plugin interactively /plugin install # View installed plugins /plugin list

Method 2: Direct Installation

# Add a plugin marketplace /plugin marketplace add https://github.com/your-org/plugin-marketplace # Install specific plugin /plugin install plugin-name # Enable/disable plugins /plugin enable plugin-name /plugin disable plugin-name

Creating Your First Plugin

Step-by-Step Plugin Creation

# 1. Create plugin directory mkdir my-first-plugin cd my-first-plugin # 2. Create plugin manifest mkdir .claude-plugin cat > .claude-plugin/plugin.json << 'EOF' { "name": "my-first-plugin", "version": "1.0.0", "description": "My first Claude Code plugin", "author": "Your Name" } EOF # 3. Add a custom command mkdir commands cat > commands/hello.md << 'EOF' Say hello to the user and explain what this plugin does. EOF # 4. Test your plugin locally /plugin install ./my-first-plugin # 5. Use your new command /hello

Real-World Plugin Examples

🔍 Code Review Plugin

Includes:

  • /review command
  • Pre-commit hook for linting
  • Code reviewer agent

🚀 Deployment Plugin

Includes:

  • /deploy command
  • Pre-deploy checklist
  • Deployment agent
  • Cloud provider MCP

🧪 Testing Plugin

Includes:

  • /test command
  • Test generator agent
  • Post-commit test hook

📊 Analytics Plugin

Includes:

  • /report command
  • Database MCP server
  • Analytics agent

Team Plugin Workflows

Organization-Wide Plugin Distribution

# 1. Create organization plugin marketplace # Host on GitHub, GitLab, or internal server # 2. Team members add marketplace /plugin marketplace add https://github.com/acme-corp/claude-plugins # 3. Install company standard plugins /plugin install acme-code-standards /plugin install acme-deployment /plugin install acme-security # 4. Everyone uses the same workflows! /review # Company standard code review /deploy # Company deployment process /security # Company security checks
Pro Tip: Plugins are the best way to standardize workflows across your team. Create organization-specific plugins that encode your team's best practices, coding standards, and deployment procedures. Everyone stays in sync automatically!

Plugin Best Practices

📝 Document Everything

Include a README.md explaining what your plugin does and how to use it

🔢 Version Properly

Use semantic versioning (1.0.0) so teams know when updates are breaking

🧪 Test Components

Test each command, agent, and hook individually before packaging

📦 Keep It Focused

Each plugin should solve one problem well, not everything

🔒 Security Review

Review all code in plugins, especially those from third parties

🔄 Update Regularly

Keep plugins updated with latest Claude Code features

🔗 Learn More About Plugins

For detailed technical documentation, see:

Advanced Workflows

🔄 Automated Code Migration

"Migrate all class components to functional components with hooks. Create a todo list for each component, and we'll work through them systematically." # Claude will: # 1. Find all class components # 2. Create a migration plan # 3. Convert each component # 4. Test functionality # 5. Update tests

🎯 Multi-Repository Operations

"I have 3 microservices. Update the API client in all of them to use the new authentication method." # Work through each repo systematically

📊 Automated Reporting

"Generate a weekly progress report by: 1. Analyzing git commits from the past week 2. Querying our PostgreSQL database for user metrics 3. Pulling task completion data from Jira 4. Creating a formatted markdown report" # Combines MCP, file operations, and data analysis

🏗️ Project Scaffolding

"Create a new microservice with: - Express.js backend - PostgreSQL database schema - Docker configuration - GitHub Actions CI/CD - Comprehensive tests - OpenAPI documentation Follow our company's standard project structure." # Claude can scaffold entire projects!

Power User Tips

Tip 1: Context Files
Create a .claude/CLAUDE.md file with project-specific instructions that Claude will always follow. Perfect for coding standards, architectural decisions, and team preferences.
Tip 2: Combine Tools
"Use our PostgreSQL MCP to analyze user behavior, then update the dashboard components to display the insights." - Claude can chain operations seamlessly.
Tip 3: Agent Delegation
For complex tasks, Claude can spawn sub-agents to work in parallel. "Research the best approach for implementing real-time notifications while I continue working on the auth system."
Tip 4: Batch Operations
"Apply this refactoring pattern to all files in the /services directory" - Claude can handle bulk operations efficiently.
Tip 5: Learning from Examples
"Look at how we implemented caching in the user service, then apply the same pattern to the product service" - Claude learns from your codebase.
Expert Level Unlocked! You now have the knowledge to use Claude Code at its full potential. Check out the "Use Cases" tab for inspiration on what to build next!

💡 You've Mastered Advanced Features!

You learned: MCPs, hooks, custom commands, agents, plan mode, and automation patterns.

See what you can build!

Use Cases

20+ real-world examples and inspiration

Real-World Use Cases

🏗️ Software Development Use Cases

1. Understanding a New Codebase

Scenario: You've joined a new team and need to understand a large, unfamiliar codebase.

Workflow

"Help me understand this codebase. Start with the overall architecture." "Show me how the authentication flow works." "Where is the payment processing logic?" "Create a diagram showing how the main components interact." "What are the main dependencies and why were they chosen?"

Result: Comprehensive understanding in hours instead of weeks.

2. Building a Complete Feature

Scenario: Add a complex new feature from scratch.

Example: Real-Time Chat Feature

"Let's build a real-time chat feature. Start with a plan." # Claude creates a detailed implementation plan "Great! Let's implement the WebSocket server." "Now create the React chat component." "Add message persistence to the database." "Implement read receipts and typing indicators." "Write comprehensive tests for the chat system." "Add documentation for the chat API."

Result: Complete, production-ready feature with tests and docs.

3. Debugging Complex Issues

Scenario: Intermittent bug that's hard to reproduce.

Example: Memory Leak Investigation

"Our Node.js service has a memory leak. Memory usage grows over time. Let's investigate systematically." # Claude will: # 1. Look for common memory leak patterns # 2. Check for event listener accumulation # 3. Review closure usage # 4. Examine cache implementations # 5. Check for unclosed resources # 6. Suggest adding heap snapshots # 7. Implement fixes

Result: Root cause identified and fixed with preventive measures.

4. Legacy Code Modernization

Scenario: Modernize old codebase to current standards.

Example: Modernizing a jQuery App

"This project uses jQuery. Let's plan a migration to React. Start by analyzing the current structure." # Claude creates migration strategy: # 1. Map jQuery components to React equivalents # 2. Create plan to incrementally replace # 3. Set up React infrastructure # 4. Migrate component by component # 5. Maintain functionality throughout # 6. Update tests as we go

Result: Systematic, safe migration to modern framework.

5. Performance Optimization

Scenario: Application is slow and needs optimization.

Example: React App Performance

"Our React app is slow. Run a performance audit." # Claude will: # 1. Check for unnecessary re-renders # 2. Look for missing memoization # 3. Analyze bundle size # 4. Check for synchronous operations blocking UI # 5. Review data fetching patterns # 6. Implement optimizations # 7. Measure improvements

Result: Significantly faster application with measurable improvements.

6. API Development

Scenario: Build a complete REST API from scratch.

Example: E-commerce API

"Build a REST API for an e-commerce platform with: - User authentication - Product catalog - Shopping cart - Order management - Payment processing Include OpenAPI documentation and comprehensive tests." # Claude will create a complete, production-ready API

Result: Complete API with auth, tests, docs, and error handling.

📊 Data Analysis Use Cases

7. Data Pipeline Creation

Scenario: Build automated data processing pipeline.

Example: User Analytics Pipeline

"Create a data pipeline that: 1. Pulls user event data from PostgreSQL 2. Aggregates daily metrics 3. Calculates key KPIs (DAU, retention, conversion) 4. Generates visualizations 5. Exports to CSV for stakeholders 6. Emails summary to team" # Uses MCPs, data processing, and automation

Result: Automated daily analytics reporting.

8. Data Cleaning & Transformation

Scenario: Clean messy data from multiple sources.

Example: Merge Customer Data

"I have customer data in 3 CSVs with inconsistent formats. Clean, deduplicate, and merge into a single normalized dataset." # Claude will: # 1. Analyze the data structure # 2. Identify issues (missing data, duplicates, format issues) # 3. Create cleaning script # 4. Normalize formats # 5. Deduplicate based on smart matching # 6. Generate clean output # 7. Provide summary stats

Result: Clean, normalized data ready for analysis.

9. Exploratory Data Analysis

Scenario: Understand a new dataset.

Example: Sales Data Analysis

"Analyze our sales data. Show me: - Overall trends - Seasonal patterns - Top products - Geographic distribution - Customer segments - Anomalies or outliers Create visualizations for each insight."

Result: Comprehensive analysis with actionable insights.

10. SQL Query Generation & Optimization

Scenario: Complex database queries needed.

Example: Business Intelligence Queries

"Using our PostgreSQL database: 1. Find users who made purchases in the last 30 days but not the previous 30 2. Calculate customer lifetime value by cohort 3. Identify products frequently bought together 4. Show me the queries and optimize them for performance" # Uses PostgreSQL MCP to query and analyze

Result: Optimized queries with performance insights.

📝 Documentation & Reporting

11. Comprehensive Documentation

Scenario: Project has minimal documentation.

Automated Documentation Generation

"Generate complete documentation for this project: - README with quickstart - Architecture overview - API documentation - Component documentation - Setup instructions - Deployment guide - Contributing guidelines - Code comments where missing"

Result: Professional, comprehensive documentation suite.

12. Technical Report Generation

Scenario: Need to report on project status to stakeholders.

Executive Summary Creation

"Generate a technical report covering: - What we built this sprint (from git commits) - Performance metrics (from database) - Bug trends (from issue tracker) - Code quality metrics - Next sprint planning Format as a professional markdown document." # Combines git analysis, MCP data, and code analysis

Result: Professional report ready for stakeholders.

13. Knowledge Base Creation

Scenario: Create internal documentation from tribal knowledge.

Example: Internal Wiki

"Create an internal knowledge base covering: - How to set up the development environment - Common troubleshooting issues - Deployment procedures - Architecture decision records - Best practices and patterns we follow Extract information from code, comments, and commit messages."

Result: Comprehensive onboarding documentation.

🧪 Testing & Quality Assurance

14. Test Suite Generation

Scenario: Code has insufficient test coverage.

Comprehensive Testing

"Our authentication module has no tests. Create: - Unit tests for all functions - Integration tests for the auth flow - Edge case coverage - Error handling tests - Mock external dependencies Aim for 100% code coverage."

Result: Complete test suite with high coverage.

15. Automated Testing Workflow

Scenario: Set up continuous testing.

CI/CD Test Integration

"Set up automated testing: - GitHub Actions workflow - Run tests on every PR - Generate coverage reports - Block merges if coverage drops - Run E2E tests on staging - Performance regression tests"

Result: Automated testing pipeline ensuring quality.

🔧 DevOps & Infrastructure

16. Docker & Kubernetes Setup

Scenario: Containerize application.

Complete Containerization

"Containerize this application: - Create optimized Dockerfile - Multi-stage builds for production - docker-compose for local development - Kubernetes manifests for deployment - Helm charts - CI/CD pipeline for container builds"

Result: Production-ready containerized application.

17. Infrastructure as Code

Scenario: Define infrastructure programmatically.

Terraform Setup

"Create Terraform configuration for our AWS infrastructure: - VPC and networking - ECS cluster for containers - RDS for database - S3 for storage - CloudFront for CDN - Route53 for DNS Include all security groups and IAM roles."

Result: Complete IaC setup for reproducible infrastructure.

🎨 Creative & Non-Traditional Uses

18. Configuration Management

"Audit all configuration files. Ensure: - No secrets in code - Consistent environment variables - Proper validation - Documentation for each config option"

19. Dependency Analysis

"Analyze our dependencies: - Which ones are actually used? - Any security vulnerabilities? - License compatibility check - Size impact on bundle - Suggest alternatives for large deps"

20. Code Archaeology

"This function is complex and has no documentation. Analyze the git history to understand: - Why was it written this way? - What problem does it solve? - Who were the main contributors? - Has it had many bug fixes? Then document it properly."
Inspiration Unlocked! These are just examples—the possibilities are endless. What will you build with Claude Code?

✨ You've Explored 20+ Use Cases!

You discovered: Coding workflows, data analysis, content creation, learning, and automation possibilities.

Keep learning and growing

Resources

Documentation, courses, videos, and community links

Resources & Learning

📚 Official Documentation

Claude Code Docs

Complete official documentation

Visit Documentation

Getting Started Guide

Official quickstart tutorial

Read Guide

MCP Documentation

Model Context Protocol guide

Learn MCP

Hooks Reference

Complete hooks documentation

Explore Hooks

Slash Commands Guide

Creating custom commands

Learn More

Best Practices

Tips from the Claude team

Read Tips

🎥 Video & Interactive Learning

📚 Interactive Course

Claude Code in Action

Official Anthropic course with hands-on labs covering architecture, implementation techniques, and MCP integration.

✓ Self-paced • ✓ Certificate available

Start Course

🎓 DeepLearning.AI

Claude Code: A Highly Agentic Coding Assistant

Free course with Anthropic's Elie Schoppik. Learn fundamentals, context management, and advanced patterns.

✓ Beginner-friendly • ✓ 2-3 hours

Take Course

📺 YouTube Channel

@anthropic-ai

Latest conference talks, product demos, and technical deep dives from the Anthropic team.

✓ New content regularly

Subscribe

🎬 Code with Claude 2025

Anthropic's first developer conference

Watch talks from Anthropic engineers covering:

  • Building production agents with MCP
  • Advanced Claude Code workflows
  • Real-world implementation patterns
  • Tool use & API best practices
Watch Conference Talks

📖 Documentation

Official Claude Code Docs

Complete reference covering installation, configuration, tools, MCPs, hooks, and troubleshooting.

✓ Searchable • ✓ Code examples

Browse Docs

⚡ Best Practices

Engineering Blog Post

Tips and tricks from Anthropic engineers for using Claude Code across different codebases and languages.

✓ Real examples • ✓ 10 min read

Read Post

📄 Essential Reading for Teams

Scaling Agentic Coding Across Your Organization

Comprehensive guide from Anthropic covering:

  • Change management strategies for AI adoption
  • Security and governance considerations
  • Measuring ROI and productivity gains
  • Best practices from enterprise deployments

Perfect for engineering leaders, CTOs, and team leads

Download Free PDF Guide

🔧 MCP Servers

MCP Server Repository

Official collection of MCP servers

Browse Servers

Database MCPs

PostgreSQL, MySQL, SQLite, MongoDB

Database MCPs

Cloud Provider MCPs

AWS, GCP, Azure integrations

Cloud MCPs

Create Your Own MCP

Build custom MCP servers

Build Guide

💡 Example Projects

Starter Templates

  • React + TypeScript: Modern React app with best practices
  • Node.js API: REST API with Express and PostgreSQL
  • Python Data Science: Jupyter notebooks with pandas and visualization
  • Fullstack Next.js: Complete Next.js app with authentication

Ask Claude: "Create a [template name] project with all best practices"

Example Workflows

  • Daily Standup Automation: Generate standup notes from git commits
  • Code Review Checklist: Automated code review based on team standards
  • Performance Dashboard: Track and visualize app performance metrics
  • Documentation Generator: Auto-generate docs from code

🎯 Quick Reference

Common Commands

# Start Claude Code claude # Get help /help # Clear conversation /clear # Run custom slash command /your-command # Exit /exit

File Locations

# Global config ~/.claude/config.json ~/.claude/CLAUDE.md # Project-specific config .claude/config.json .claude/CLAUDE.md .claude/commands/ # MCP configuration ~/.mcp.json

Useful Prompts

  • "Explain how this codebase works"
  • "Find and fix this bug: [description]"
  • "Refactor [function/module] to be more efficient"
  • "Add comprehensive tests for [feature]"
  • "Generate documentation for this project"
  • "Review my changes and suggest improvements"
  • "Help me debug why [issue] is happening"

🤝 Community & Support

GitHub Issues

Report bugs and request features

Submit Issue

Anthropic Discord

Join the community

Join Discord

Support Docs

Get help with common issues

Visit Support

Status Page

Check service status

View Status

📖 Recommended Reading

Anthropic Blog Posts

External Resources

  • Prompt Engineering Guide: Learn effective prompting techniques
  • AI-Assisted Development: Patterns and practices
  • Code Quality with AI: Maintaining standards

🚀 Get Started Now

Ready to dive in? Pick a starting point:

Read Docs Browse MCPs

🏆 Congratulations! You've Completed the Guide!

You've explored everything from basic commands to advanced automation. You now have all the knowledge you need to use Claude Code effectively.

Now go build something amazing! 🚀

Start Using Claude Code

🔧 Troubleshooting & FAQ

Running into issues? You're not alone! Here are solutions to the most common problems users encounter.

🚀 Getting Started Issues

❌ Claude won't start / "Command not found"

Problem: When I type claude, I get "command not found"

Solutions:

  1. Verify installation: which claude
  2. Reinstall: npm install -g @anthropic-ai/claude-code
  3. Check your PATH environment variable
  4. Try with npx: npx @anthropic-ai/claude-code
  5. Restart your terminal after installation

🔑 API Key / Authentication Issues

Problem: "Authentication failed" or "Invalid API key"

Solutions:

  1. Verify key is set: echo $ANTHROPIC_API_KEY
  2. Set key: export ANTHROPIC_API_KEY=your-key
  3. Add to your shell profile (~/.zshrc or ~/.bashrc) to make it permanent
  4. Check key is valid in Anthropic Console
  5. Regenerate key if needed

🔒 Permission Errors

Problem: "Permission denied" when reading/writing files

Solutions:

  1. Check file permissions: ls -la filename
  2. Make sure you're running Claude Code from the correct directory
  3. Check that .gitignore isn't blocking the files
  4. Verify you have write access to the project directory
  5. Don't use sudo unless absolutely necessary (not recommended)

💬 Usage Issues

🤐 Claude Isn't Responding

Problem: Claude started but isn't responding to my messages

Solutions:

  1. Check your internet connection
  2. Verify you haven't hit API rate limits
  3. Try /clear to reset the conversation
  4. Restart Claude Code completely
  5. Check Anthropic status page
  6. Look for error messages in your terminal

🐛 Claude Made a Mistake

Problem: The code Claude wrote has bugs or isn't what I wanted

Solutions:

  1. Be more specific in your request and provide more context
  2. Ask Claude to fix it: "This has a bug in line 42 where..."
  3. Iterate: "Can you try a different approach?"
  4. Provide examples of what you want
  5. Undo with git: git restore <file>
  6. Ask Claude to explain its reasoning: "Why did you choose this approach?"

↩️ How Do I Undo Changes?

Problem: I want to revert Claude's changes

Solutions:

  1. Single file: git restore <file>
  2. All changes: git restore .
  3. Hard reset: git reset --hard HEAD (careful!)
  4. Pro tip: Always review changes before accepting
  5. Pro tip: Work in a branch: git checkout -b feature-name
  6. Pro tip: Commit frequently so you can easily revert

🐌 Performance Is Slow

Problem: Claude is taking a long time to respond

Solutions:

  1. Large codebases (>100k lines) may be slower—this is normal
  2. Be specific to reduce scope: "Fix the login function" vs "Fix everything"
  3. Check your internet connection speed
  4. Try smaller, focused questions
  5. Use agents for long-running tasks that can work in the background
  6. Break complex tasks into smaller steps

📁 Claude Can't Find My Files

Problem: Claude says it can't find a file that exists

Solutions:

  1. Check the file path is correct: ls path/to/file
  2. Make sure you're in the right directory: pwd
  3. Check if the file is in .gitignore (Claude respects .gitignore)
  4. Try using the full path instead of a relative path
  5. Ask Claude to search: "Find all files named X"

⚙️ Configuration Issues

🔌 MCPs Not Working

Problem: My MCP server isn't connecting

Solutions:

  1. Check ~/.mcp.json formatting (must be valid JSON)
  2. Verify the MCP server is installed: npm list -g
  3. Check logs for error messages
  4. Try a simple test MCP first (like filesystem)
  5. Restart Claude Code after changing MCP config
  6. Verify the server path is correct

🪝 Hooks Not Firing

Problem: My hooks aren't running

Solutions:

  1. Check .claude/config.json syntax (must be valid JSON)
  2. Verify hook commands work independently in terminal
  3. Check file permissions on hook scripts: chmod +x script.sh
  4. Look for error messages in Claude Code output
  5. Try a simpler hook first to test the system
  6. Make sure the hook path is absolute or relative to project root

⌘ Custom Commands Not Working

Problem: My custom slash command doesn't show up

Solutions:

  1. Check command file location: .claude/commands/commandname.md
  2. Verify file name matches: /commandname → commandname.md
  3. Restart Claude Code after adding new commands
  4. Check file has content (can't be empty)
  5. Try /help to see if your command is listed

🔌 Plugins Not Working

Problem: My installed plugin isn't working or showing up

Solutions:

  1. Verify plugin is installed: /plugin list
  2. Check plugin is enabled: /plugin enable plugin-name
  3. Restart Claude Code after installing plugins
  4. Verify plugin marketplace URL is correct: /plugin marketplace
  5. Check plugin manifest (.claude-plugin/plugin.json) is valid JSON
  6. Try reinstalling: /plugin uninstall plugin-name then reinstall
  7. Check plugin documentation for specific requirements

🤔 Common Questions

💰 How much does Claude Code cost?

Claude Code uses your Anthropic API key, which is billed based on usage. You can check pricing at anthropic.com/pricing. You only pay for what you use.

🔐 Is my code secure?

Your code is sent to Anthropic's API for processing. Anthropic does not train models on your API data. Review Anthropic's privacy policy for full details.

🌐 Can I use Claude Code offline?

No, Claude Code requires an internet connection to communicate with Anthropic's API. All the AI processing happens in the cloud.

👥 Can multiple people use Claude Code on the same project?

Yes! Each person needs their own API key. Claude Code works great in team environments. Just make sure to use git branches to avoid conflicts.

📦 What languages/frameworks does Claude Code support?

Claude Code works with all programming languages and frameworks. It's especially good with popular languages like Python, JavaScript/TypeScript, Java, Go, Rust, and more.

🆚 Claude Code vs claude.ai - What's the difference?

Claude Code: CLI tool integrated with your terminal, has direct file system access, can run commands, optimized for development workflows.

claude.ai: Web interface, great for general conversations and non-coding tasks, requires copy-pasting code.

Use Claude Code for software development, use claude.ai for other tasks.

🆘 Still Stuck?

Can't find your issue? Here's how to get help:

  1. Search Documentation: Official Claude Code Docs
  2. Check GitHub Issues: Known issues and feature requests
  3. Ask the Community: Anthropic Discord
  4. Contact Support: Anthropic Support

When asking for help, include:

  • The exact error message (copy the full text)
  • What you were trying to do
  • Your operating system and version
  • Claude Code version: claude --version
  • Steps to reproduce the issue
Most Issues Are Quick to Solve! The majority of problems have simple solutions. Check the common issues above first—you'll probably find your answer. If not, the community is here to help!