OpenAI Electron

Command line tool to generate Electron application with OpenAI API

Installation

You can install this project by using NPM

npm install --global openai-electron-cli

You can check if the project installed properly and the version of the project

openai-electron --version

You can see all the avaiable commands

openai-electron -H

Windows User Issue

If you are Windows user and get error like this:

openai-electron : File C:\Users\xxxx\AppData\Roaming\npm\openai-electron.ps1 cannot be loaded because running scripts
is disabled on this system. For more information, see about_Execution_Policies at
https:/go.microsoft.com/fwlink/?LinkID=135170.
At line:1 char:1
+ openai-electron
+ ~~~~~~~~~~~~~~~
    + CategoryInfo          : SecurityError: (:) [], PSSecurityException
    + FullyQualifiedErrorId : UnauthorizedAccess

You can fix it by running this command (it may cause a security issue)

Set-ExecutionPolicy -Scope CurrentUser -ExecutionPolicy Unrestricted

Command Usage

install

Installing all electron depencies that needed to run an Electron application

set-prompt

Set the prompt that will send to OpenAI to generate the source code. Example:

openai-electron set-prompt "write a simple blog in HTML with CSS"

get-prompt

Get the prompt that given by user. Example:

openai-electron get-prompt
# output:
# write a simple blog in HTML with CSS

generate-code

Generate the source code by sending the prompt to OpenAI API. You need to have OpenAI API Key. Recommend to read How to create OpenAI API keys. Example:

openai-electron generate-code --openai-key "sk-4EdVbFhGP..."
# output:
# ✅ Source code generated successfully
# ✅ Source code saved and application is ready to run

edit-code

View and edit the generated source code from OpenAI with text editor (default: vim). Example:

openai-electron edit-code --text-editor vim

run

Run the generated source code as desktop application via electron

run-webserver

Run the generated source code as Web application via HTTP server. Example:

openai-electron run-webserver
# output:
# ╭────────────────────────────────────────────────╮
# │                                                │
# │    Application run on http://localhost:5000    │
# │                                                │
# ╰────────────────────────────────────────────────╯

export

Export the generated source code as executable program via electron-forge. Example:

openai-electron export -o .
# output:
# ? What is your platform? Linux
# ✅ Application packaged successfully
tree -L 3 .
# output:
# .
# └── out
#     ├── electron_app-linux-x64
#     │   ├── LICENSE
#     │   ├── LICENSES.chromium.html
#     │   ├── chrome-sandbox
#     │   ├── chrome_100_percent.pak
#     │   ├── chrome_200_percent.pak
#     │   ├── crashpad_handler
#     │   ├── electron_app
#     │   ├── icudtl.dat
#     │   ├── libEGL.so
#     │   ├── libGLESv2.so
#     │   ├── libffmpeg.so
#     │   ├── libvk_swiftshader.so
#     │   ├── locales
#     │   ├── resources
#     │   ├── resources.pak
#     │   ├── snapshot_blob.bin
#     │   ├── swiftshader
#     │   ├── v8_context_snapshot.bin
#     │   ├── version
#     │   └── vk_swiftshader_icd.json
#     └── make
#         ├── electron-app_1.0.0_amd64.deb
#         └── electron_app-1.0.0.x86_64.rpm

# 6 directories, 19 files