How to Build an LLM-Powered CLI Tool in Python

This tutorial shows how to build a real-time, AI-powered command line tool in Python using OpenAI’s Realtime API. You’ll create llm-explain, a CLI utility that explains any shell command by streaming LLM responses directly into your terminal. The guide covers setting up a WebSocket client, handling streaming output, and extending the tool with optional “AI agent” capabilities like tool-calling and safe shell execution. By the end, you’ll have a reusable framework for building your own AI-native CLI assistants.