run export API_KEY="[your api key]"

run commandchat chat" to chat with Chatgpt and the default chat model is gpt-3.5-turbo

run commandchat setting to modify the chatting model, FYI, the input should be formatted as json

• eg: switch to model text-davinci-003,

{
  "modelName": "text-davinci-003",
  "apiUrl": "https://api.openai.com/v1/completions"
}

• pre-preparing
  • docker
• mock data is saved in the folder stub, if you want to add a new request, you can add a new file in its sub-folder mappings. And the response of a new request is saved in the sub-folder __files
• modify the configuration setting.json, let the endpoint connect to the mocker server as follows:

  {
    "modelName": "gpt-3.5-turbo",
    "apiUrl": "http://localhost:8080/v1/chat/completions"
  }

• run startstub.sh to start the mocker server, you will see the following output in your console

  2023-04-12 06:50:22.626 Verbose logging enabled
  2023-04-12 06:50:24.321 Verbose logging enabled
  /$$      /$$ /$$                     /$$      /$$                     /$$
  | $$  /$ | $$|__/                    | $$$    /$$$                    | $$
  | $$ /$$$| $$ /$$  /$$$$$$   /$$$$$$ | $$$$  /$$$$  /$$$$$$   /$$$$$$$| $$   /$$
  | $$/$$ $$ $$| $$ /$$__  $$ /$$__  $$| $$ $$/$$ $$ /$$__  $$ /$$_____/| $$  /$$/
  | $$$$_  $$$$| $$| $$  \__/| $$$$$$$$| $$  $$$| $$| $$  \ $$| $$      | $$$$$$/
  | $$$/ \  $$$| $$| $$      | $$_____/| $$\  $ | $$| $$  | $$| $$      | $$_  $$
  | $$/   \  $$| $$| $$      |  $$$$$$$| $$ \/  | $$|  $$$$$$/|  $$$$$$$| $$ \  $$
  |__/     \__/|__/|__/       \_______/|__/     |__/ \______/  \_______/|__/  \__/
  
  port:                         8080
  enable-browser-proxying:      false
  disable-banner:               false
  no-request-journal:           false
  verbose:                      true
  

• Then you can run go run . chat to debug your app, you will see the verbose log in the console.