Claude Code¶

Documentation for Claude Code, an AI-powered development tool integrated into this workflow.

Overview¶

Claude Code provides AI-assisted development capabilities for code generation, analysis, and problem-solving. This setup uses AWS Bedrock for model inference, enabling local-first Claude Code execution.

Setup¶

1. Configure AWS Credentials¶

Add your AWS credentials to ~/.aws/credentials:

[bedrock]
aws_access_key_id = XXXXXXXXXXX
aws_secret_access_key = YYYYYYYYYYYYYYYYY

2. Configure AWS Region¶

Add the Bedrock profile to ~/.aws/config:

[profile bedrock]
region = us-west-2

3. Install Claude Code¶

Install Claude Code globally via npm:

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

4. Create Wrapper Script¶

Create a wrapper script at ~/bin/claude to simplify usage and model selection:

#! /bin/bash

export CLAUDE_CODE_USE_BEDROCK=1

export AWS_REGION=us-west-2
export AWS_PROFILE=bedrock
export ANTHROPIC_MODEL="anthropic.claude-sonnet-4-5-20250929-v1:0"
export ANTHROPIC_SMALL_FAST_MODEL="us.anthropic.claude-haiku-4-5-20251001-v1:0"

mymodel=${ANTHROPIC_SMALL_FAST_MODEL}
if [[ "$1" == "opus" ]]; then
  mymodel=$1
elif [[ "$1" == "sonnet" ]]; then
  mymodel="anthropic.claude-sonnet-4-5-20250929-v1:0"
fi

~/.claude/local/claude --dangerously-skip-permissions --model $mymodel $@

5. Launch Claude Code to Set Up PATH and Permissions¶

Instead of manually configuring PATH and permissions, let Claude Code do it for you:

mkdir -p ~/bin
bash ~/bin/claude

When you launch this, Claude Code will:

Ask you to configure your environment - Claude will detect that ~/bin/claude needs to be made executable and added to PATH
Provide setup commands - Follow Claude's suggestions to:
Make ~/bin/claude executable: chmod +x ~/bin/claude
Add ~/bin to your PATH by updating .bashrc or .zshrc
Reload your shell - Execute the reload command Claude suggests

6. Test Claude Code¶

Once Claude has helped you set up the environment, launch Claude Code normally:

# Create a test directory
mkdir -p ~/test-project
cd ~/test-project
git init

# Launch Claude Code with default Haiku model
claude .

When Claude Code launches for the first time, it will: 1. Ask for your SSH keychain passphrase (cached for 4 hours) 2. Initialize the Claude Code environment 3. Start the interactive session in the git repository

Important: Git Repository Requirement¶

Claude Code must always be initialized inside a git repository. This is a requirement for the tool to function properly.

Setting Up a Git Repository¶

You have two options:

Option 1: Create a Local Repository¶

mkdir my-project
cd my-project
git init

Then run Claude Code inside this directory:

claude .

Option 2: Create a Repository on GitHub¶

Visit github.com/new to create a new repository
Clone it to your local machine:

git clone https://github.com/YOUR_USERNAME/my-project.git
cd my-project

Run Claude Code:

claude .

Model Comparison¶

Understanding the differences between the three available models helps you choose the right tool for each task:

Aspect	Haiku	Sonnet	Opus
Speed	⚡ Fast	⚡⚡ Medium	🐌 Slow
Cost	💰 $0.80/MTok	💰💰 $3.00/MTok	💰💰💰 $15.00/MTok
Performance	Excellent	Very Good	Superior
Complex Tasks	⭐	⭐⭐⭐ Medium	⭐⭐⭐⭐⭐ High
Coding Skills	⭐⭐ Medium	⭐⭐⭐ Good	⭐⭐⭐⭐⭐ Excellent

Usage¶

Default (Haiku - Fast)¶

Run Claude Code with the default fast Haiku model for quick fixes and simple tasks:

claude /path/to/project

Best for: - Quick bug fixes - Simple code generation - Refactoring small sections - Cost-conscious tasks

Sonnet (Balanced)¶

Run with the balanced Sonnet model for most development work:

claude sonnet /path/to/project

Best for: - Complex feature development - Detailed code analysis - Medium-complexity problems - Good balance of speed and capability

Opus (Most Capable)¶

Run with the most capable Opus model for challenging problems:

claude opus /path/to/project

Best for: - Difficult architectural decisions - Complex problem-solving - Full-project analysis - Maximum capability required

Configuration Notes¶

Bedrock Integration: Uses AWS Bedrock for model inference
Context Window: Defaults to 1M context window for Sonnet
No Confirmation: Configured to skip permission prompts for streamlined workflow
Model Selection: Easy switching between Haiku (fast), Sonnet (balanced), and Opus (capable)

Tips and Best Practices¶

Use Haiku for quick fixes and simple tasks
Use Sonnet for complex development and analysis
Use Opus for challenging problems requiring maximum capability
Ensure AWS credentials are properly configured before use
The wrapper script automatically handles model selection and AWS environment setup