Since I published The Whetstone - a piece about encoding your own practice into AI so the work stays distinctively yours instead of converging on the model’s defaults -the same question has landed in my inbox dozens of times. How do you actually do this? What does the surfacing look like? How do I get from “I have a practice” to “I have a set of skill files that extend it”?
I’ve tried to answer it. In voice notes. In half-written replies. In one truly painful attempt at writing instructions.
None of it worked. And the more I tried, the more I understood why.
The whole point of the Whetstone is that it’s yours. Your references, your signals, the way your mind connects dots, the practices you’ve developed over years and never had to look at directly. I can’t surface that for you.
What I can do is codify the process - the provocations that get you out of your own blind spots, the dimensional scaffolding that stops you fishing in one part of the pond, the sequence of questions that takes a vague practice and makes it deployable.
So that’s what I’ve built.
The Whetstone is a guided process - created in Lovable, with other tools layered in around it - that runs you through the same arc I use myself. Surfacing. Interrogation. Naming. Encoding. By the end, you walk away with at least eight to ten v1 skill files: deep articulations of your own practice, written in your voice, ready to plug into Claude, ChatGPT, Gemini, or whichever AI you work in. Yours, properly.
You go through it once, end-to-end, going deep on the handful of areas where your tacit knowledge actually lives. After that, the bot drops out. You take the files and develop them from there.
It won’t write your strategy for you. The skill files won’t replace the work - instead they put you to work, sparring with your own mind, mining your own memory, being your own devil’s advocate. The friction is the point. What you walk out with is unmistakably yours: a more precise version of your own thinking, instead of a more generic version of someone else’s.
It takes time. A good couple of hours minimum, probably longer, and I wouldn’t advise rushing it in one sitting - better to split it across two or three sessions and let things surface between them. The files keep developing as you do. They’re never finished, because you’re never finished.
It'll be my first paid product. The IP and the build are mine; the price will reflect that. And it’ll be released in three to four weeks. Maybe. I’m building it in the five spare minutes I have between everything else, and one of the reasons I’m putting this out now is honestly to give myself a kick up the ass to finish it properly.
The accountability of strangers waiting on something is the most reliable production schedule I know!
If you want a heads up when it opens, you can sign up to the waitlist here.
More soon.
Fascinating - really looking forward to trying this, it does make me wonder (having read your original posts on this) whether what you are getting towards codifying is fundamentally different between different people or if all of this is pointing to something more universally shared. Its something I've been thinking about a lot recently and I've been using discernment type practices (in my case learned from Quakerism) and meditation to explore. My hunch is that this looks different depending on the type or extent of neurodivergence but that we are pointing to something universal about life force and how it manifests through the human brain.
Encoding your practice is the first step to scaling a soul-led brand. Such a valuable framework for any visionary leader!!