GPT-inputAssistant

An application designed to allow querying OpenAI's GPT from any input field.

What can the app do?

• Initiate LLM calls without disconnecting from the workspace.
• Rapid LLM calling.

Features

• Multilingual support [en, zh, jp, es, az]
• Cross-platform compatibility [win, mac]
• Ability to customize prompts

How to Use

Getting Started

1. Open GPT-inputAssistant click "Set API KEY" to set your OpenAI Key.
2. Copy the desired text to your clipboard
3. Press shift + space to query GPT
4. Press ESC key to halt generation
5. Triple click ESC key to quickly clear context

Defining User HotKeys

Click "Set API KEY" and GPT-inputAssistant will open an env.txt file. Simply add a new line to the file like this GPT_HOTKEYS=space+shift, save and close the file. The keycode references can be found here!

Importing Prompts

Simply copy the json format below:

{
  "name": "ChatGPT",
  "model": "gpt-3.5-turbo-0613",
  "headMessages": [
    {
      "role": "system",
      "content": "You are a helpful assistant."
    }
  ],
  "maxContext": 20
}

{
  "name": "Translate into Chinese",
  "model": "gpt-3.5-turbo-0613",
  "headMessages": [
    {
      "role": "system",
      "content": "You're a translator translating any text I give you into Chinese. Just return the result, no need for explanations."
    }
  ],
  "maxContext": 1
}

Then select the application's import menu.

For more templates, go to the prompts folder in this repository.

DEMO

The demos can be found here and here

Building

Use the build_win.bat command for Windows or the build.sh command for other platforms.

Credit

Applications utilized in the development of this app:

https://github.com/getlantern/systray

https://github.com/go-vgo/robotgo

https://github.com/robotn/gohook

https://github.com/hanyuancheung/gpt-go

New Owners: VagAnts