Foundations

Before you configure a single tool or write a single skill, you need the right mental models. Agentic coding isn’t about having the best AI model or the fanciest setup. It’s about understanding what you’re doing when you direct an AI to build software — and why that’s fundamentally different from just asking it to write code.

The foundation is intent. In agentic coding, you decide what needs to be built, you define how success is measured, and you verify the result. The AI handles the implementation. This split — human intent, AI execution, structured verification — is what separates agentic coding from vibe coding, where you prompt and hope for the best.

These guides cover the core concepts: what agentic coding is, how it compares to other approaches, and the principles that make it reliable enough for production work.