inlinai

AI based cli to run commands generated by human text prompt.

Features

• Prompt-based Command Generation: The tool asks for a prompt and generates shell commands based on that prompt.
• Run Commands: After generating the commands, the tool asks if you want to execute them in your terminal.
• Hugging Face Integration: The tool uses the Hugging Face API and the mistralai/Mistral-7B-Instruct-v0.3 model to generate relevant shell commands.

Installation

1. Clone this repository:

   git clone https://github.com/UsboKirishima/inlinai
   cd inlinai

2. Install dependencies:

   npm install

3. Create a .env file with your Hugging Face API key:

   HF_API_KEY=<your-hugging-face-api-key>

   Note: The other entries (OPENAI_KEY, OPENAI_ENDPOINT, OPENAI_MODEL, DEEPSEEK_API_KEY) are experiments and can be ignored for this tool.

Usage

1. Run using npm/pnpm/yarn:

   npm run start gen "read file README.md"
   pnpm start gen "read file README.md"
   yarn start gen "read file README.md"
   

2. Run by installing global package:

   npm i --global @333revenge/inlineai
   yarn global add @333revenge/inlineai
   pnpm add -g @333revenge/inlineai

   after installing run command:

   inlineai gen "read file README.md"
   

Screenshots

1. Get kernel version:

2. Print "pizza" ten times: