Open WebUI Assistant is a companion app that allows you to insert text generated by your Open WebUI instance anywhere on your computer using keyboard shortcuts, and offers a user interface for setting up the server URL and JWT token. The two primary keyboard shortcuts are:

1. Cmd+Option+i: Use this shortcut when the text input field is active to request text completion from Open WebUI Assistant.
2. Cmd+Option+o: This command uses the normal chat API endpoint for interaction with your Open WebUI instance, not opening a separate chat window.

Before installing Open WebUI Assistant, make sure you have make, node, and npm installed on your system.

If you don't have make or node installed, install them using your system package manager or refer to the official documentation and Node.js download page.

1. Clone Open WebUI Assistant's GitHub repository:

   git clone https://github.com/open-webui/assistant.git
   cd assistant

2. Run make to prepare the build environment:

   make

3. Install Open WebUI Assistant using the following command:

   make install

4. Configure Open WebUI Assistant:

   • After installation, find and run the Open WebUI Assistant companion app located in your user's ~/Applications directory.
   • In the settings window, enter your Open WebUI server URL and the JWT token you can get from Settings > Account in the WebUI. The companion app's UI is just a setting interface for these two values.

5. Use Open WebUI Assistant:

   • After configuring Open WebUI Assistant, use the keyboard shortcuts Cmd+Option+i to request text completion or Cmd+Option+o for interaction with your Open WebUI instance using the normal chat API endpoint.

For more information, troubleshooting, or feature requests, visit Open WebUI Assistant's GitHub repository or say hello in our Discord server.