Run your store with agents
Learn to let an AI agent do your store's daily work safely, with you in the cockpit. A short course shows, on a real store, the same method the book teaches: the loop, the safeguards, backup and rollback, and the autonomy dial. Leave your email and you start right away.
What you'll be able to do
After the course you no longer do a catalog update, a product description or a campaign stage by hand. You let the agent propose, read the plan, approve, and let it out only when it's right. The same daily work in a fraction of the time, and you know exactly what was written to the store.
“Think of a fast, literal-minded new hire on their first week: one who will do exactly what you asked, at astonishing speed, including the parts you didn't mean.”
Visual, not a lecture
The course is built in the same console look as this page: you watch the agent work, you don't sit through slides. Below is the slot for a real slide from it.

Six things, on a real store
Explore, plan, confirm, execute, report. You see one turn of it in a real store, on screen doing the work, not as a slide deck.
Platform access and permissions so you don't hand over everything at once. What the agent gets to do, and what it never does.
The agent fetches the changing fields before the write and can write them back. The rollback is walked once, not left to memory.
Back up before the write, approve before it's live, verify after, a known rollback path. This is what keeps the store intact.
When the agent may run on its own and when you read every batch. You learn to set the dial by a task's risk, not by mood.
A concrete small, safe job for your own store that you can hand the agent today and check yourself.
the agent runs, you spot-check
the agent proposes, you read every batch
the agent sketches, you drive
The method came out of a near-miss
The course teaches the method I built after I nearly wrecked a client's live store theme. Without the method, an agent isn't automation, it's a fast way to break something.
In 2025 I set an agent a small job on a client's live store: add one feature to the theme. I wrote the prompt in a hurry and left the agent running on its own. I saw it start writing to the theme and head the wrong way. Had I let it finish, it would have rewritten the core files of a live store's theme, and I had nothing to put them back: no backup, no snapshot, no rollback path. I hit escape in time. Nothing broke. What saved that store was a reflex and some luck, not a method.
That morning is why this exists. The loop, the backup rules, the confirm gate and the rollback path are the safety layer I built afterward, so it could never again come down to luck and reflex.
If you run the store yourself
The course is for the operator who wants to use agents to get work done, not just to experiment. No coding background needed. If you can manage — be clear about the outcome and check the work — you already have the hard part.
A taster, not the whole book
Honestly: the course is a fast, visual taster of the method. It's enough for a first win. The book (€49, toolkit included) teaches it in full — 22 chapters across every store function. The course shows you whether to go deeper. Leave your email and we'll begin.
Common questions
Nothing. You leave your email and get the lessons. Unsubscribe anytime.
No catch. The course is a fast, visual taster of the method. The book (€49) teaches it in full, across every store function. The course shows you whether to go deeper.
The same method, shorter: the loop, the safeguards, wiring, the autonomy dial. The book goes deeper and applies it to every store function.
No. Everything is shown in plain words, on a real store.
Short. A few lessons by email, you start with the first one. Built so you get a first win fast.
The whole method is in the book
The book (€49, toolkit included) goes deeper and applies the method to every store function. The course is the free start.