Blog
Thoughts on design systems, AI-native delivery and vibe coding.
llms.txt for Shops: Build the Signpost Once the Road Exists
An llms.txt is quick to create and just as quick to become useless. When the file is due, what it should look like for a shop, and the mistakes we see most often in our checks.
From Score to Sequence: Which Agent Readiness Gaps to Close First
A readiness score shows where you stand. It does not tell you where to start. How to sort agent readiness gaps by impact and effort, and why the order looks almost the same for every shop.
What Machine-Readable Product Data Really Means
Everyone says product data has to be machine-readable. But what does that mean in practice? Which fields an AI agent actually needs, and how to check they arrive.
Why Your Shop Is Invisible to AI Agents
Your product data is all there, and an AI agent still cannot see it. Why that happens, and how to check it yourself in two minutes.
The 6 Dimensions of Agent Readiness
Agent Readiness consists of six dimensions that work together. How to test what an agent takes from your website and whether it can act on your behalf.
AI-native Development Without a System: Three Patterns
When AI-native development works without a foundation, the same three patterns keep showing up. They turn a productive setup into an expensive exercise.
Your website is getting a second visitor
How to make your website readable and usable for AI agents.
The design system as your AI context layer
Tokens, components and documentation give AI the context it needs to deliver reliable production code.
Vibe coding is a method, not a gamble
From prompt roulette to reproducible AI-native product delivery: the system behind the speed.