When I shared the video of the personal operations dashboard I built in Claude Code, I didn’t expect it to reach as many people as it has. The most common request since then has been some version of the same thing: don’t just show me, help me get my own.
A few weeks ago I released the scope document — the full spec Claude and I built before writing a line of code. That gave people a real starting point. But I was honest about its limitation: building from the scope still means many hours of work alongside Claude Code, iterating through the inevitable bugs, before you have something running.
So here’s the next step. I’m making the actual code available.
This is a testing release. I’m inviting paid subscribers to download the repository, deploy it, and try it on their own. It’s a work in progress, shared in the spirit of “come test this with me,” not “here’s a polished product.”
I’m not turning the dashboard into a product you can sign up for and pay to use. I’ve said this before: it solves a very specific problem for a specific kind of person, and the moment I tried to generalize it for everyone, I’d be rebuilding Notion or ClickUp. Tools that already exist and are excellent at being general.
Sharing the repo isn’t a step toward a SaaS. It’s the opposite. It’s me handing you the thing itself so you can run it, own it, and shape it, instead of renting it from me.
Support is limited. I can’t offer hands-on installation help. But paid subscribers can chat with me, and I’ll do my best to answer questions as they come up.
The repository fast-tracks everything the scope document asked you to build by hand. Instead of starting from a spec and a blank project, you start from working code.
But here’s the caveat you need to understand before you deploy it: this installs the system as I have it configured. All the tools I currently use come with it: the Content Manager, the specific domain structure, the health tracking, everything. Some of that will be irrelevant to you. If you deploy it as-is, you’ll get a working version of my life management system, not yours.
So this is my genuine recommendation, and it’s the best of both paths:
Take the repo download into a new Claude Code project. Give Claude the code, the original scope document, and the design files. Then customize it to fit your own life.
That way you get a working foundation instead of a blank page, but you shape it around how you think, what you do, and the areas of your life that actually matter. You keep the parts that fit, strip the parts that don’t, and add what’s missing. That’s the difference between inheriting my system and building your own on top of a head start.
This isn’t a one-click install, and I want you to go in with realistic expectations. To run this you’ll need:
Some comfort with development, or a real willingness to work through problems with Claude Code
A Supabase account for the database
API keys (OpenAI for transcription, Anthropic for parsing)
Somewhere to host a Node.js project, such as XCloud
Google Calendar API access if you want the calendar integration
If that list looks approachable, you’re in a good position to test this. If it looks intimidating, the scope document plus Claude Code might actually be the gentler on-ramp, because you’ll build understanding as you go rather than inheriting a system you didn’t assemble.
Just so it’s clear where this fits:
The scope document — build it from scratch with Claude Code. Most work, most understanding, fully yours from the first line.
The repository (this release) — start from my working code. Fastest path to something running, but requires customization to make it truly yours.
The recommended path — repo + scope + design files, dropped into Claude Code, customized to you. The head start of the code with the flexibility of a rebuild.
Comments
Nothing yet. Say the first thing.
Sign in to join the conversation.