What Naetive is, and what it is not

6 September 2026

When something is new, people reach for the nearest familiar box to put it in. Naetive gets put in a few, and most of them are wrong. So here is the plain version. What it is, and what it is not.

What it is

Naetive gives your project a living understanding that every AI agent inherits. The decisions, the constraints, the reasons behind the code, captured as you work, grounded in the real files, and carried forward so the next agent you open already knows them.

That is the whole thing. Understanding that belongs to the project, not to the chat that happened to discover it.

It is not a memory tool

This is the most common mix-up. Memory tools store what was said, then try to recall the relevant bits later. That is a log of conversation.

Understanding is different. It is a current, structured model of what is true about the project, kept aligned with the code. Not "what did we talk about", but "what is this project, why is it this way, and what is still open". A log grows forever and drifts. Understanding is meant to stay correct.

It is not a wiki or a docs folder

A wiki is understanding written by hand, on a schedule, by someone with the discipline to keep it current. Which is why wikis rot. The gap between the docs and the code widens every week and nobody trusts them.

Naetive is captured as you work, not written up afterwards, and every claim is tied to the code it came from. When the code moves, the claim is flagged for review instead of silently going stale. It is the opposite of a doc you have to remember to update. A CLAUDE.md sits somewhere in between, and is worth keeping for what it is genuinely good at.

It is not an orchestration tool

Naetive does not run your agents, schedule them, or wire them together. Plenty of tools want to be the conductor. Naetive is not that.

It sits underneath whatever agents you already use, connected over MCP. Claude, Claude Code, ChatGPT, Codex. It gives them one shared understanding to work from. You keep your tools and your workflow. They just stop working blind.

It is not the model, and it is not more context

A better model does not fix this, and neither does a bigger context window. Both still open cold on everything that is not in the repo. Understanding is the layer above the model, which is why no model release touches it.

So: not memory, not a wiki, not orchestration, not a smarter model. A living understanding, native to your project, that every agent inherits.