Claude Code

Overview

Claude Code is a powerful command-line interface designed specifically for interacting with large language models directly from your terminal. It enables seamless LLM interactions in your daily development workflow.

Key Features

• Deep code understanding & editing: Claude Code excels at understanding complex codebases, allowing it to assist with code generation, refactoring, and debugging tasks.
• Works where developers already work: Integrates directly into your terminal and IDE, enabling you to leverage LLM capabilities without disrupting your existing workflow.
• Full development workflow support: From writing and testing code to documentation and deployment, Claude Code supports the entire software development lifecycle.
• Configurability & control: Customize prompts, set context windows, and manage conversation history to tailor the experience to your specific needs.
• Multi-platform support: Compatible with various operating systems and development environments.

Use Cases

• Code generation and modification
• Debugging and bug fixing
• Automating routine development tasks
• Prototyping new features
• Writing and running tests
• Integrating with existing tools and workflows
• Creating pull requests from issues
• Making cross-file and large scope edits
• Learning and onboarding support

Setup Instructions

Get Claude Code up and running on your machine with these simple steps.

1. Install Claude Code:

   Native Install (Recommended)

   macOS, Linux, WSL:

   curl -fsSL https://claude.ai/install.sh | bash

   Windows PowerShell:

   irm https://claude.ai/install.ps1 | iex

   Windows CMD:

   curl -fsSL https://claude.ai/install.cmd -o install.cmd && install.cmd && del install.cmd

   Homebrew

   If you have Homebrew installed:

   brew install --cask claude-code

   NPM

   If you have Node.js 18 or newer installed:

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

2. Connect to relaxAI:

   export ANTHROPIC_API_KEY="rak_apikey" >> ~/.zshrc
   export ANTHROPIC_BASE_URL="https://api.relax.ai" >> ~/.zshrc
   export API_TIMEOUT_MS="300000" >> ~/.zshrc
   export ANTHROPIC_DEFAULT_HAIKU_MODEL="GLM-46" >> ~/.zshrc
   export ANTHROPIC_DEFAULT_SONNET_MODEL="GLM-46" >> ~/.zshrc
   export ANTHROPIC_DEFAULT_OPUS_MODEL="GLM-46" >> ~/.zshrc
   export CLAUDE_CODE_DISABLE_NONESSENTIAL_TRAFFIC=1 >> ~/.zshrc
   export CLAUDE_CODE_DISABLE_ANALYTICS=1 >> ~/.zshrc
   export CLAUDE_CODE_DEFAULT_MODEL="GLM-46" >> ~/.zshrc

   Then, reload your shell configuration:

   source ~/.zshrc

   ANTHROPIC_DEFAULT_HAIKU_MODEL

   Used for fast and lightweight tasks that don’t require deep reasoning:

   • Listing files and directories
   • Reading simple file content
   • Executing system commands
   • Basic operations that need speed over complexity

   ANTHROPIC_DEFAULT_SONNET_MODEL

   The main workhorse model for most coding tasks:

   • Writing and editing code
   • Reasoning about architecture
   • Debugging and problem-solving
   • Analyzing existing code
   • Tasks requiring balance between speed and capability

   ANTHROPIC_DEFAULT_OPUS_MODEL

   Used for the most complex tasks requiring maximum reasoning:

   • Large code refactorings
   • Complex architectural problems
   • Tasks requiring multiple reasoning steps
   • Debugging particularly difficult issues

3. Verify installation:

   claude-code --version

4. Testing the setup:

   Change into a directory with code and run:

   claude --model GLM-46 "Explain the code in this directory"

Common workflows

Learn about common workflows with Claude Code. Each task includes clear instructions, example commands, and best practices to help you get the most from Claude Code with relaxAI.

Understand new codebases

Get a quick codebase overview:

When you join a new project and need to understand its structure quickly, follow these steps:

1. Navigate to the project root directory:

   cd /path/to/project

2. Start Claude Code:

   claude --model GLM-46

3. Ask for a high-level overview:

   > give me an overview of this codebase

4. Dive deeper into specific components:

   > explain the main architecture patterns used here

   > what are the key data models?

   > how is authentication handled?

   Tip

   Start with broad questions, then narrow down to specific areas. Ask about coding conventions and patterns used in the project, and request a glossary of project-specific terms.

Find relevant code:

To locate code related to a specific feature or functionality:

1. Ask Claude to find relevant files:

   > find the files that handle user authentication

2. Get context on how components interact:

   > how do these authentication files work together?

3. Understand the execution flow:

   > trace the login process from front-end to database

   Tip

   Be specific about what you’re looking for and use domain language from the project.

Fix bugs efficiently

When you encounter an error message and need to find and fix its source:

1. Share the error with Claude:

   > I'm seeing an error when I run npm test

2. Ask for fix recommendations:

   > suggest a few ways to fix the @ts-ignore in user.ts

3. Apply the fix:

   > update user.ts to add the null check you suggested

   Tip

   Tell Claude the command to reproduce the issue and get a stack trace. Mention any steps to reproduce the error, and let Claude know if the error is intermittent or consistent.

Refactor code

Update old code to use modern patterns and practices:

1. Identify legacy code for refactoring:

   > find deprecated API usage in our codebase

2. Get refactoring recommendations:

   > suggest how to refactor utils.js to use modern JavaScript features

3. Apply the changes safely:

   > refactor utils.js to use ES2024 features while maintaining the same behavior

4. Verify the refactoring:

   > run tests for the refactored code

   Tip

   Ask Claude to explain the benefits of the modern approach, request that changes maintain backward compatibility when needed, and do refactoring in small, testable increments.

Work with tests

Add tests for uncovered code:

1. Identify untested code:

   > find functions in NotificationsService.swift that are not covered by tests

2. Generate test scaffolding:

   > add tests for the notification service

3. Add meaningful test cases:

   > add test cases for edge conditions in the notification service

4. Run and verify tests:

   > run the new tests and fix any failures

Claude can generate tests that follow your project’s existing patterns and conventions. When asking for tests, be specific about what behavior you want to verify. For comprehensive coverage, ask Claude to identify edge cases you might have missed.

Create pull requests

Create well-documented pull requests for your changes:

1. Summarize your changes:

   > summarize the changes I've made to the authentication module

2. Generate a pull request with Claude:

   > create a pr

3. Review and refine:

   > enhance the PR description with more context about the security improvements

4. Add testing details:

   > add information about how these changes were tested

   Tip

   Ask Claude directly to make a PR for you. Review Claude’s generated PR before submitting, and ask Claude to highlight potential risks or considerations.

Handle documentation

Add or update documentation for your code:

1. Identify undocumented code:

   > find functions without proper JSDoc comments in the auth module

2. Generate documentation:

   > add JSDoc comments to the undocumented functions in auth.js

3. Review and enhance:

   > improve the generated documentation with more context and examples

4. Verify documentation:

   > check if the documentation follows our project standards

   Tip

   Specify the documentation style you want (JSDoc, docstrings, etc.), ask for examples in the documentation, and request documentation for public APIs, interfaces, and complex logic.

Work with images

Work with images in your codebase and get Claude’s help analyzing image content:

1. Add an image to the conversation using any of these methods:

   • Drag and drop an image into the Claude Code window
   • Copy an image and paste it into the CLI with ctrl+v
   • Provide an image path to Claude: “Analyze this image: /path/to/your/image.png”

2. Ask Claude to analyze the image:

   > What does this image show?

   > Describe the UI elements in this screenshot

   > Are there any problematic elements in this diagram?

3. Use images for context:

   > Here's a screenshot of the error. What's causing it?

   > This is our current database schema. How should we modify it for the new feature?

4. Get code suggestions from visual content:

   > Generate CSS to match this design mockup

   > What HTML structure would recreate this component?

   Tip

   Use images when text descriptions would be unclear or cumbersome. Include screenshots of errors, UI designs, or diagrams for better context. You can work with multiple images in a conversation.

Reference files and directories

Use @ to quickly include files or directories without waiting for Claude to read them:

1. Reference a single file:

   > Explain the logic in @src/utils/auth.js

   This includes the full content of the file in the conversation.

2. Reference a directory:

   > What's the structure of @src/components?

   This provides a directory listing with file information.

3. Reference multiple files:

   > Compare @file1.js and @file2.js

   Tip

   Use @file-path to include specific files and @directory-path to include directory listings. This saves time and ensures Claude has the exact context needed.

Resume previous conversations

Resume a previous session when starting Claude Code:

• claude --continue --model GLM-46 continues the most recent conversation in the current directory
• claude --resume --model GLM-46 opens a conversation picker or resumes by name

From inside an active session, use /resume to switch to a different conversation.

Name your sessions:

1. Name the current session using /rename:

   > /rename auth-refactor

2. Resume by name later from the command line:

   claude --resume auth-refactor --model GLM-46

   Or from inside an active session:

   > /resume auth-refactor

   Tip

   Name sessions early when starting work on a distinct task. Use --continue for quick access to your most recent conversation. Use --resume session-name when you know which session you need.

Use Claude as a unix-style utility

Add Claude to your verification process:

Add Claude to your build script:

{
    "scripts": {
        "lint:claude": "claude --model GLM-46 -p 'you are a linter. please look at the changes vs. main and report any issues related to typos. report the filename and line number on one line, and a description of the issue on the second line. do not return any other text.'"
    }
}

Pipe data through Claude:

cat build-error.txt | claude --model GLM-46 -p 'concisely explain the root cause of this build error' > output.txt

Control output format:

1. Use text format (default):

   cat data.txt | claude -p 'summarize this data' --output-format text > summary.txt

2. Use JSON format:

   cat code.py | claude -p 'analyze this code for bugs' --output-format json > analysis.json

3. Use streaming JSON format:

   cat log.txt | claude -p 'parse this log file for errors' --output-format stream-json

   Tip

   Use --output-format text for simple integrations, --output-format json when you need the full conversation log, and --output-format stream-json for real-time output.

Create custom slash commands

Create reusable slash commands for your project that all team members can use.

Create project-specific commands:

1. Create a commands directory in your project:

   mkdir -p .claude/commands

2. Create a Markdown file for each command:

   echo "Analyze the performance of this code and suggest three specific optimizations:" > .claude/commands/optimize.md

3. Use your custom command in Claude Code:

   > /optimize

Add command arguments with $ARGUMENTS:

1. Create a command file with the $ARGUMENTS placeholder:

   echo 'Find and fix issue #$ARGUMENTS. Follow these steps: 1. Understand the issue 2. Locate the relevant code 3. Implement a solution 4. Add tests 5. Prepare a PR description' > .claude/commands/fix-issue.md

2. Use the command with an issue number:

   > /fix-issue 123

   This replaces $ARGUMENTS with “123” in the prompt.

   Tip

   Command names are derived from the filename (e.g., optimize.md becomes /optimize). Project commands are available to everyone who clones the repository. The $ARGUMENTS placeholder can be positioned anywhere in your command template.

Create personal slash commands:

1. Create a commands directory in your home folder:

   mkdir -p ~/.claude/commands

2. Create a Markdown file for each command:

   echo "Review this code for security vulnerabilities, focusing on:" > ~/.claude/commands/security-review.md

3. Use your personal custom command:

   > /security-review

   Tip

   Personal commands work across all your projects and are only available to you, not shared with your team.

Best Practices

• Be specific in your prompts: Provide clear context and detailed requirements for better results
• Use incremental changes: Make small, testable changes rather than large refactors all at once
• Leverage session naming: Name your sessions with descriptive names early to find them easily later
• Combine with existing tools: Integrate Claude Code into your CI/CD pipeline and shell scripts
• Review generated code: Always review and test code changes before committing
• Use project commands: Share common workflows with your team through .claude/commands/
• Ask for explanations: Request explanations of changes to understand the reasoning behind suggestions

Resources

• relaxAI API Documentation
• relaxAI Pricing
• Authentication Guide
• Making Requests