Complete Guide to Installing and Configuring Claude Code: A Detailed Tutorial for Setting Up Your AI Programming Environment

Introduction

Claude Code is a command-line AI programming tool developed by Anthropic. Thanks to its excellent code generation and project comprehension capabilities, it's becoming the go-to tool for an increasing number of developers. However, due to network restrictions, users in China often encounter various obstacles during installation and usage. Based on a tutorial shared by the Bilibili creator "跟着娟姐学Python" (Learn Python with Sister Juan), this article provides a detailed walkthrough of the complete Claude Code installation and configuration process. It also covers how to connect to domestic large language models like DeepSeek via CC Switch, allowing you to experience AI-powered programming without needing a VPN.

Step 1: Install the Node.js Runtime Environment

Claude Code is built on Node.js, so you'll first need to install the Node.js runtime environment locally.

Downloading and Installing Node.js

1. Visit the Node.js official website (nodejs.org) and select the Windows version
2. Click "Windows Installer" to download the installation package
3. Double-click the downloaded installer to launch the setup wizard
4. Click "Next" on the welcome screen and accept the license agreement
5. Choose the installation path (it's recommended to install on a non-system drive, such as the D drive)
6. Keep the default settings for subsequent steps, check the necessary options, and click "Install"
7. Click "Finish" once the installation is complete

Tip: It's recommended to download the LTS (Long-Term Support) version for better stability. If you see an option to "Automatically install necessary tools" during installation, check it — this will automatically configure environment variables and other required settings.

Step 2: Install Claude Code via npm

Running the Installation Command

Once Node.js is installed, you can install Claude Code globally using the npm command:

1. Press Win + R, type CMD, and open the command prompt
2. Enter the following installation command and press Enter:

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

3. Wait for the installation process to complete

Verifying the Installation

After installation, verify it by entering the following command in the command prompt:

claude --version

If you see a version number in the output, Claude Code has been installed successfully.

Step 3: Configure Login Bypass

Due to network restrictions, you may not be able to complete the Anthropic account login when using Claude Code directly. You'll need to modify a configuration file to bypass the login verification step.

Steps to Modify the Configuration File

1. Locate the configuration file at: C:\Users\{your-username}\.claude\claude-code.json
2. Open the file with a text editor (such as Notepad or VS Code)
3. Add the appropriate configuration code to the file (to skip login verification)
4. Save the file, then re-enter claude in the command prompt to launch

At this point, Claude Code will start normally but will show a "not logged in" status. Don't worry — in the next step, we'll use a third-party tool to connect to domestically available large language models.

Step 4: Install CC Switch to Connect to Domestic LLMs

CC Switch is a model-switching tool designed specifically for Claude Code. It supports replacing the underlying model with domestic LLMs like DeepSeek, solving the problem of Chinese users being unable to directly access the Claude API.

Downloading and Installing CC Switch

1. Visit the official CC Switch GitHub page
2. Scroll to the bottom of the page and find the download link for the Windows version
3. Download the installer and double-click to install
4. Choose the installation path and click "Install" to complete the setup
5. A desktop shortcut will appear after installation

Configuring the DeepSeek Model

1. Double-click the desktop shortcut to open CC Switch
2. Click the "+" icon in the upper right corner and select "Add Provider"
3. Select DeepSeek as the provider
4. Obtain a DeepSeek API Key:
   • Go to the DeepSeek Open Platform (platform.deepseek.com)
   • Log in and navigate to the "API Keys" page
   • Click "Create API Key," enter a name, and generate it
   • Copy the generated key (make sure to save it securely — it cannot be viewed again after closing)
5. Paste the API Key into the corresponding field in CC Switch
6. Click "Add" to complete the configuration

Enabling the Model

After adding the model, you need to manually click the "Enable" button to activate the DeepSeek model. Newly added models are disabled by default, so make sure the toggle has been switched to the enabled state.

Step 5: Switch Models in Claude Code

After completing the CC Switch configuration, you need to restart Claude Code for the new model to take effect:

1. Close the current command prompt window
2. Press Win + R again → type CMD to open a new command prompt
3. Type claude to launch the program
4. On the startup screen, select Enter (press E and hit Enter)
5. Type the /model command to open the model switching menu
6. Select DeepSeek V3 (or another configured model) from the model list
7. Press Enter to confirm — the model switch is complete

After completing these steps, you can start using Claude Code powered by DeepSeek for AI-assisted programming.

Usage Tips and Important Notes

Model Selection Recommendations

• DeepSeek V3: Ideal for everyday programming tasks with fast response times and excellent cost-effectiveness
• If you need stronger reasoning capabilities, consider connecting other supported models through CC Switch

Common Troubleshooting

• npm installation fails: Check your network connection, or try using a domestic npm mirror (such as the Taobao mirror) to speed up downloads
• Configuration file not found: Make sure you've run the claude command at least once — the configuration file is automatically generated on the first run
• No response after switching models: Verify that the API Key is entered correctly and confirm that your DeepSeek account has sufficient balance
• CC Switch doesn't detect Claude Code: Try reinstalling CC Switch and ensure Claude Code is properly installed in the global path

Cost Considerations

DeepSeek's API charges are based on token usage, but the overall pricing is quite affordable. It's recommended to start with a small top-up (e.g., 10 RMB) for testing. Once everything is confirmed to be working properly, top up based on your actual usage.

Conclusion

By following these five steps — installing the Node.js environment, deploying Claude Code, modifying the configuration file, installing and configuring CC Switch, and switching models — users in China can successfully set up a complete Claude Code AI programming environment. Although the underlying model has been swapped from Claude to DeepSeek, Claude Code's excellent command-line interface and workflow remain fully intact. Combined with DeepSeek's powerful code generation capabilities, this setup can fully meet everyday development needs. As domestic LLMs continue to improve, the practical value of this combined solution will only grow.