# Claude Code

### Your intelligent coding partner


<br>


## Overview

Claude Code is an AI assistant designed to help you write, understand, and improve code. It works alongside you, learning from your project to provide contextual assistance throughout your development workflow.


<br>
<br>


## What It Does

**Understands**
Reads and explains your codebase, making complex architectures accessible.

**Creates**
Writes new features based on your requirements and existing patterns.

**Refines**
Identifies and resolves issues while maintaining code quality.

**Tests**
Validates functionality to ensure reliability.

**Adapts**
Works autonomously on extended tasks while respecting your preferences.


<br>
<br>


## Getting Started


### Configure Your Environment

Create a `claude.md` file in your project root. This serves as Claude's guide to understanding your workflow.

```
Include your development practices
Define coding standards
Explain business-specific terminology
Outline architectural decisions
```

Use `/init` to generate a starter file, then refine it to match your needs.


<br>


### Choose Your Mode

**Plan Mode**
Claude develops a strategy, creates a task list, and waits for your approval.
`Shift + Tab` to toggle

**Auto Mode**
Claude works independently, making changes as it proceeds.
Best for routine tasks with clear requirements.


<br>


### Enable Extended Thinking

Give Claude additional reasoning time for complex problems.
`Tab` to toggle

Particularly effective when combined with Plan Mode for architectural decisions.


<br>
<br>


## Essential Features


### Checkpoints

Your work is automatically preserved before each significant change. If something goes wrong, simply press `Escape` twice to return to a previous state.


<br>


### Subagents

Specialized assistants for focused tasks. Each maintains its own context, keeping your main workflow clean and efficient.

Create subagents for:
- Build validation
- Test execution
- Code review
- Architecture planning

Configure via `/agents`


<br>


### Hooks

Automated responses to specific events in your workflow.

Common uses:
- Format code after edits
- Notify when tasks complete
- Validate changes before execution
- Maintain coding standards


<br>


### Custom Commands

Create shortcuts for frequent tasks.

`/review` instead of "Please review this code for bugs and suggest improvements"

Define your own vocabulary for common workflows.


<br>
<br>


## Best Practices


### Discovery

Let Claude explore your codebase to understand structure and patterns. Particularly valuable when joining new projects or working with unfamiliar components.


<br>


### Planning

Use Plan Mode with Extended Thinking for complex features. Provide design documents when available. Allow Claude to identify edge cases and suggest improvements.


<br>


### Implementation

Write tests first. Claude's success rate improves dramatically when it can validate its work against defined criteria.

Without tests: Variable results
With tests: Consistently high quality


<br>


### Context Management

Claude performs best with focused, relevant information.

- Use `/compaction` to summarize long conversations
- Deploy subagents for specialized tasks
- Be selective about file reads
- Allow Claude to maintain its own notes for complex work


<br>
<br>


## Common Situations


### Claude agrees too readily

Add instructions to your `claude.md` encouraging critical feedback. Request that it challenge assumptions and identify potential issues.


<br>


### Incomplete understanding

Spend more time in Plan Mode. Provide additional business context. Share design documents for complex features.


<br>


### Excessive file creation

Redirect when noticed. This behavior continues to improve with each model iteration.


<br>


### Avoiding difficult problems

Use Extended Thinking. Provide encouragement and direction. Break complex problems into smaller steps.


<br>
<br>


## Quick Reference

```
/init          Create claude.md file
/compaction    Compress conversation
/agents        Configure subagents
/memory        Add persistent context

Shift + Tab    Toggle Plan Mode
Tab            Toggle Extended Thinking
Escape × 2     Rewind changes
```


<br>
<br>


## Integration


### Terminal
Command-line interface for direct interaction

### VS Code
Native sidebar integration with your editor

### JetBrains IDEs
Terminal-based with editor awareness

### GitHub Actions
Automated code review on pull requests

### Custom Applications
Embed via Claude Agents SDK


<br>
<br>


## Current Release

### Claude 4.5 Sonnet

Enhanced reasoning and contextual understanding. Improved performance on long-running tasks. More reliable debugging capabilities.

Recent additions include checkpoint support, plugin system, and native VS Code integration.


<br>
<br>


## Learn More

Documentation
[docs.anthropic.com/en/docs/claude-code](https://docs.anthropic.com/en/docs/claude-code/overview)

Common Workflows
[docs.anthropic.com/en/docs/claude-code/common-workflows](https://docs.anthropic.com/en/docs/claude-code/common-workflows)

Updates
GitHub changelog and release notes


<br>
<br>


## Philosophy

Start simple. Build gradually. Write tests. Plan carefully. Review thoroughly.

Claude is your partner in development. You guide the direction. Claude handles the details.


<br>
<br>


---

*Based on Claude 301 Training by Anthropic Applied AI*