GitHub Copilot is a AI assistant developed by GitHub to help developers write code. It is a Visual Studio Code plugin based on OpenAI Codex, providing features such as code suggestions, auto-completion, auto-fixing, and auto-refactoring. Currently, GitHub Copilot also supports chat functionality, powered by GPT-4. However, it is only available within the context of VS Code. In this article, I will explain how to set up a chat service using GitHub Copilot, allowing you to use it anywhere.

Open-Source Components

• ChatGPT-Next-Web: A well-designed, cross-platform ChatGPT web UI with support for GPT3, GPT4, and Gemini Pro.
• copilot-gpt4-service
• An AWS EC2 server is needed to run the copilot-gpt4-service, but you can choose any cloud service provider you prefer or run it locally.
• Vercel: A free static website hosting service to deploy the ChatGPT-Next-Web service.
• Cloudflare: A free CDN service used for domain name resolution.

Obtaining the GitHub Copilot Token

Please refer to this document, GitHub Copilot Token. On a Mac, you can use the following command to obtain the token:

1

bash -c "$(curl -fsSL https://raw.githubusercontent.com/aaamoon/copilot-gpt4-service/master/shells/get_copilot_token.sh)"

As shown in the screenshot, you need to copy the token to the clipboard for later use.

Running the copilot-gpt4-service

1. The copilot-gpt4-service supports Docker deployment, and the official project provides a docker-compose file. You only need to modify the environment variables.

1. Add a server block in NGINX:

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20

server {
     listen 443 ssl;
     listen [::]:443 ssl;
     server_name chat.example.com;
 ​
     ssl_certificate /etc/nginx/cert/public.pem;
     ssl_certificate_key /etc/nginx/cert/private.key;
 ​
     location / {
         proxy_pass http://127.0.0.1:8086/;  # Replace with the copilot-gpt4-service address
         rewrite ^/(.*)$ /$1 break;
         proxy_redirect off;
         proxy_set_header Host $host;
         proxy_set_header X-Forwarded-Proto $scheme;
         proxy_set_header X-Real-IP $remote_addr;
         proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
         proxy_set_header Upgrade-Insecure-Requests 1;
         proxy_set_header X-Forwarded-Proto https;
     }
 }

Don’t forget to reload the NGINX configuration file.

1. Add a domain name resolution for “chat.example.com” on Cloudflare, pointing to the IP address of your NGINX server.

Deploying the ChatGPT-Next-Web Service on Vercel

ChatGPT-Next-Web supports one-click deployment to Vercel. You don’t need to make any changes.

Once the deployment on Vercel is complete, you can add a custom domain name, such as “chatnext.example.com”.

Testing

Open your browser and enter “chatnext.example.com”. Click on the settings button in the bottom left corner and select “Custom Interface”. Enter " https://chat.example.com" as the API endpoint, and use the GitHub Copilot token copied in the first step as the API token. Click on “Save”, and you can start chatting.