Memory keeps what. Unl keeps why.
Models change. Sessions reset. Your decisions shouldn’t. Why keeping the reasons behind your decisions takes an agent, not a memory. Your AI already knows how to reason. Unl gives it your reason.
A no without a why
is a yes.
Tell a model “don’t do that” and it can’t tell which neighbouring cases you meant, so it treats them as open. Give it the reason and it can work out the cases you never wrote down.
So Unl doesn’t steer your AI with stop signs. It steers with your why: a frame of reasons, as broad or as fine as you want, that keeps the model pointed where you meant it to go.
Memory keeps what.
Unl keeps why.
Memory keeps what you said. A prompt file holds what you told it. Neither keeps why. Unl keeps what you decided and why, what replaced it and what still holds.
What you said
What you told it
What you decided, and why
What you use today,
and what Unl adds.
| The usual way | Where it stops | With Unl |
|---|---|---|
| Custom instructions in ChatGPT or Claude | Stays inside one app, and applies to every chat whether it's relevant or not | The same decisions reach every AI you use, and only when they bear on the turn |
| CLAUDE.md, .cursorrules, project knowledge | Lives in one repo and one editor, kept up by hand, and goes stale | Your decisions follow you across repos, tools and models, and update the moment you change your mind |
| Memory features | Remember what you said, not why you decided it | Keeps each decision with its reason, and what it replaced |
| Building your own MCP memory server | You host it, configure it and keep it running | Connect once. Unl is the why agent already running |
You stop writing things down for your AI. Unl notices decisions as you make them, and hands each AI the one that matters, when it matters.
What keeping why
actually requires.
Memory is one capability. Keeping why faithfully sets a whole chain of requirements that memory does not meet.
That is why Unl is an agent, not a memory.
Memory accumulates.
Why compounds.
A fact is a point. An instruction is a limit. A why has direction. Tell a model exports run through the queue and it follows the rule. Tell it why, because inline exports failed on large accounts, and it can reason about the case you never wrote down.
Your decisions don’t live alone. The reasons behind product, architecture, customers and pricing bear on one another. Unl brings the relevant path into the model, so a fresh AI starts from the direction you already set, not from zero.
The model doesn’t just inherit answers. It inherits the direction they came from.
Plain answers.
01Is Unl a memory tool?
No. Unl is your why agent. Memory keeps what was said. Unl keeps why: what you decided, the reason behind it, whether it still holds, and what replaced it. Your AI works from your why, not from a transcript.
02Does Unl change what I decided?
Never on its own. When something you say changes a decision, your AI can offer to update it, and nothing moves until you say yes.
03Does it work with Claude Code, Cursor and ChatGPT?
Yes. One address over MCP, the same on every surface. Once it is added and you have signed in, it is on in every conversation, chat or coding, with nothing to say or paste.
04What does it cost?
Free to start. Paid plans are Basic $10, Pro $30 and Max $90 a month, each a weekly usage allowance calibrated to the AI plan you already pay for, and the API is on every tier. What you keep stays readable on every plan.
05Where does my why live?
In your own workspace, tied to your sign-in. Any AI you connect reads the relevant part when it bears. Nothing reaches into your model or your tools. Your why stays yours, kept as long as you want it, gone when you say so.