12 terminal windows. A prompt in the wrong one.
This is how I taught my AI to stop me.
Earlier this year I was working on multiple projects in parallel. Each in its own terminal window. AI agent on the left, knowledge system on the right, research project in the middle. Up to 12 windows at once.
It worked. Until it didn’t.
At some point, notes from project A ended up in the database of project B. Research results were saved in the wrong context. Facts got mixed up. And the worst part: I didn’t notice until days later. Because the AI didn’t report an error. It just did what I said. In the wrong window.
The problem wasn’t the prompt. The prompt was good. The problem was that the AI didn’t know where it was.
AI Has No Sense of Space
ChatGPT, Claude, Copilot. They all answer what you ask. But none of them asks back: Does this question even belong here? Is this the right context? Does this fit what we’re currently working on?
A human would notice. If you suddenly ask about recipes in a meeting about budget planning, someone asks: Everything okay? An AI just carries on.
The Solution: Scope Checks
So I taught the system to check its context. Every terminal window gets a scope: a project context that defines which data belongs to it. When I tell the AI agent “Save this note”, the system checks: Does the content match the current scope? If I dictate a note about a different project while in the research project, the agent stops and asks: “This doesn’t sound like the project we’re working in right now. Right window?”
It sounds simple. It was. A few rules, a comparison, a warning. No machine learning, no fine-tuning. Just the question: Does this belong here?
But almost no AI tool asks this question on its own.
The Real Insight
We talk a lot about better prompts. About prompt engineering, context windows, system instructions. But almost nobody talks about what happens when a perfect prompt lands in the wrong context. And the more windows, projects and agents you use in parallel, the more important exactly that becomes.
It’s not the quality of the answer that matters. It’s whether it lands in the right place.
The right prompt in the wrong terminal window.