AI-CMD is an innovative project designed to seamlessly integrate GPT capabilities with the Windows terminal, enabling users to directly ask questions and generate command-line code using natural language. By leveraging an OpenAI proxy, AI-CMD offers accessibility without the need for a VPN, simplifying the process of tackling terminal-related queries and automating command generation.

1. Clone the project from GitHub:

   git clone https://github.com/rhouselyn/ai-cmd
   

2. Navigate to the project directory and install the required packages:

   cd ai-cmd
   pip install -r requirements.txt
   

3. run the setup:

   python setup.py
   

4. Add the project directory to your system's PATH environment variable (or copy the dictionary in output of setup.py):
   • As for how to find the PATH environment variable, refer to this

• In config.py, input your api_key.
• You can specify your preferred model and language settings for a tailored experience.

It provide a simple process for generating commands using natural language. It doesn't provide much flexibility, but it's more convenient and faster.

• 🏃 Typing gen followed by a directive in the terminal prompts GPT to translate it into a command, which is then copied to the clipboard automatically.
• 💑 Connect multiple row operations using && or &

  

• 🎇 For potentially hazardous operations, it adds a cautionary note:

  CAUTION: This command is dangerous!

unlike gen, ask is used to ask questions with dialog context and get more responses.

• 👢 Use ask followed by your question in the terminal. AI-CMD will automatically fetch the context from the terminal, process the inquiry with GPT, and generate a response.

  

• 😝 If the response contains terminal commands, they are automatically extracted and copied to the clipboard so that you can paste directly to the terminal.

  

• 🍔 You don't need to worry about consuming too much tocken, it will only intercept instructions from back to front until it exceeds 1000 words.
• ✈️ Because of Windows don't provide any api for aproach to acquire history output in current dialog, a brief screen flash indicates the use of a virtual keyboard for copying and processing, ensuring the clipboard's content is promptly cleared afterwards.

• ⭕ If the api speed limit is reached, the program automatically retries until it is lifted。
• 🔴 Run time more than 15 seconds automatically exit the program

• ❤️ This project draws inspiration from the cli-gpt project, adopting its approach to prompt design for effective and intuitive user interactions.
• 💕 The document is mainly written by chatGPT4.