The Morning Build Scope is the third prompt in the Operator Second Brain Starter Pack. It drafts the first version of the daily morning prompt that you paste into your Operator Second Brain workspace every morning. Ten minutes to set up. One paste to use every day after.
Run this prompt once, after you have set up the eight files inside a Claude or ChatGPT Project. Re-run it once a quarter, or any time the one thing you want from AI in the morning changes.
Have three things ready: the names of the eight files in your base, a one-line description of your current top priority, and a one-line description of the single most valuable thing you want AI to tell you at the start of every day.
You are helping me draft my morning prompt for an operator second brain system.
I have a workspace with eight markdown files: Goals and Priorities, Daily Work Log, Session Log, Pending Actions, Revenue Log, Function Architecture, Brand Voice, and Working Memory.
My current top priority is:
[one line]
The single most valuable thing I want AI to tell me every morning is:
[one line]
Your job is to draft a morning prompt I can paste into Claude or ChatGPT that:
1. Instructs the model to read all eight files as context.
2. Tells me what I worked on yesterday and what is still open.
3. Names the single most important thing to do today, tied to my top priority.
4. Flags one thing I am likely to miss, based on the Pending Actions file.
5. Stays under 300 words, direct operator voice, no motivation language, no em dashes.
Return only the prompt. No explanation. I should be able to paste it directly as the custom instruction of my Project.
Paste into Claude or ChatGPT. Fill in the two bracketed lines. Copy the returned prompt and save it as the custom instruction of your Claude Project, or paste it manually every morning in ChatGPT. Run it once to verify the output. Adjust if it misses your top priority.
A strong output is a single prompt under 300 words, specific to your priority, that returns a briefing you would actually act on. If the output reads like a generic morning journaling template, it is wrong. Re-run with sharper inputs.
Letting the prompt return motivational language. Writing a top priority that is too broad to surface a daily action. Running the morning prompt once and then forgetting to update the Daily Work Log, which makes tomorrow's briefing hollow.
The morning prompt is the moment the memory base stops being storage and starts being leverage. By day five, the prompt is reading your own history back to you and telling you what to do with it. That is the point Karpathy was making. You do not have to remember everything. You just have to make sure something does.