The Unix Philosophy for Agentic Coding

Most people use AI coding agents backwards. They hand the agent a problem and ask it to solve the whole thing. The agent reads, reasons, generates, and hopes for the best. There's a better way. One that's cheaper, more predictable, and already well understood. It's the Unix philosophy , applied to how we work with AI. The Pattern The Unix philosophy boils down to: do one thing well, compose small tools, let the shell orchestrate. When you work with an AI coding agent, the agent is the shell. Here's how I think about it: Break the problem down. Don't hand the agent a big, vague goal. Decompose it into sub-problems. If a tool exists, use it. Refactoring, formatting, linting, deployment: these are solved problems. Don't ask the AI to reinvent them. If no tool exists, build one. A small, deterministic script is better than an LLM making judgment calls where none are needed. The agent orchestrates. It decides what to do, in what order, with which tools. That's where its intelligence adds va