profile

The Intelligence Briefing

One file. Two tools. Two answers.


Hello Reader,

Each week, I share one insight, one prompt, and a short grounding practice to help you grow with AI in alignment, not overwhelm.

This is for conscious entrepreneurs who value substance over shortcuts.


AI Insight That Matters πŸ’‘

It Just Answers Anyway

When an AI is missing your context, it does not stop and it does not warn you. It just answers anyway.

I learned this months ago when I created the Human Stack approach to context engineering. I keep my rules for AI in one file called AGENTS.md. How I write, what I never say, what to check before anything ships. Claude Code reads that file at the start of every session, which is one of the reasons I do my serious work there. As a check, I asked a second tool, pointed at the same folder, to quote the first line of that file back to me, word for word. It could not. It had never opened the file. And every answer it had given me sounded fine.

Nothing visibly breaks when a file gets missed. The answers keep coming, confident and well written, and they sound the same whether your context arrived or not. There is no error message for a rule that never got read. There is just an answer that sounds like every other answer.

That is why I stopped trusting how the answers sound and started running a ten-second test instead. Open a fresh session. Before you give it any work, ask it to quote the first line of your rules file, word for word. A summary can be faked. A quote cannot. Either it comes back exact and your setup is working, or it does not and now you know.

The fix, when the check fails, is usually one plain sentence. Mine was a line at the front of the folder telling every tool to read AGENTS.md in full before starting any task. No clever syntax, just words a model can act on.

And people work this way too. Someone who has not read the brief rarely says so. They nod, they start, and the work comes back polished and slightly off. Silence is not agreement, and a confident answer is not proof of understanding. When it matters, ask for the one thing that could only come from having read.

β€œ
Silence is not agreement, and a confident answer is not proof of understanding.
β€” Colin Scotland

AI Prompt That Works ✨

Turn Your Repeated Corrections Into Standing Rules

Most people's real rules are not written down anywhere. This prompt pulls them onto one page you can put in front of your AI every time.

The Prompt:

I am going to paste several pieces of work an AI produced for me, along with the corrections I made to each one.
​
Read them and find the corrections I keep repeating.
Ignore the one-offs. I only want the patterns that show up more than once.
​
Then write those patterns as a single page of standing rules, addressed to an AI as instructions it can act on.
Plain sentences, no jargon. Cover how I write, what I
never say, what to check before producing anything, and what to do when it is unsure.
​
Where a rule is clearly implied by my corrections but I have never stated it outright, write it anyway and mark it, so I can confirm it or cut it.
​
[Paste the work and your corrections here.]

Save what comes back where your AI can reach it, then start a fresh session and ask it to quote you the first rule, word for word, before you give it any work. The quote is your proof the rules are arriving, and it costs ten seconds a week to keep.


Grounding Practice ☯

The Difference Between Sure and Known

Find somewhere you will not be interrupted. Two or three minutes is plenty.

Bring to mind something you are certain about. A view you hold, an opinion you would defend without pausing for breath. Hold it there and notice what happens in your body. Where does the certainty sit, and how fast does it arrive? Notice how little it asks of you.

Now let that go and bring to mind something you know from having lived it. A thing your hands know. A thing your history taught you. Notice where that one sits instead. It usually lands lower, and slower, and it carries no urgency to be agreed with.

Stay with the difference a moment longer. Both of them feel like knowing from the inside. Only one has been anywhere.


Personal Note ❀️‍πŸ”₯

β€œPutting things off is the biggest waste of life. It snatches away each day as it comes, and denies us the present by promising the future.”
β€” Seneca, On the Shortness of Life

I am writing a book called The Amplified Coach. That sentence has been true for me for months. The book keeps getting pushed from one week to the next. It was on my list last Monday. It is on my list this Monday. In between, everything with a deadline and another person attached went first, and the book went nowhere.

When people ask how it is coming, I say it is coming.

Seneca would recognize that answer. It promises the future to avoid the present, and it is smoother than the truth. The work I care about most is the work I keep not doing, and no one is chasing me for it, which is exactly why it slips. Client work has a deadline and feels urgent, and the book just waits.

I would bet you have one of these too. The thing you would name first if someone asked what your real work is, sitting underneath a pile of things that are merely overdue. No one is chasing you for it either. That is not proof it can wait. It is the reason it has waited this long.

I do not have a tidy fix to hand you this week. I have a book that deserves better than my leftover hours, and so, probably, do you. What is yours? Reply and name it. Naming mine here is the most progress I have made on it in the past few weeks.


🎬 Go Deeper

From the blog: AI Won’t Replace You. It Will Reveal You.​

The companion piece to this week. Same principle pointed at your own work rather than your tools, because the thing AI exposes fastest is the difference between what you actually know and what you have been repeating.

On YouTube

Step-by-step walkthroughs and tactical implementation for conscious AI integration.


Thanks for being here. If this resonates, share it with someone who would benefit from hearing this.

Reply and let me know what landed for you. πŸ™

Colin

P.S. If you are building a business you can run without burning yourself out, that is the whole of the Humane Business Accelerator. The next cohort starts in September, and the preview is open now at humanebusiness.ai/accelerator. Come and have a look.

The Intelligence Briefing

Weekly field notes on scaling intelligence without losing your humanity. Get the strategic frameworks, unfiltered insights, and human-first AI methodologies.

Share this page