# FocusLM

> FocusLM builds structured, self-configuring, proactive memory for AI: an agent interviews you about the problem you are working through, designs a folder ontology for it, and files everything you feed it into that structure.

AI-Chat is the general multi-model chat inside FocusLM; a Project is the memory. They are different things and the distinction matters when reading anything below.

Project memory is available over MCP — query it from Claude Desktop or your own agents. AI-Chat itself does not read project memory.

These files are English on every domain. The product interface is localized, and AI-Chat answers in the language the user writes in — the description of the product is kept in one language on purpose, so there is one text to keep true.

## Start here

- [Full product brief](https://focuslm.pl/llms-full.txt): The canonical description, in full. Read this one first.
- [Home](https://focuslm.pl/): What the product is, for people.
- [Pricing](https://focuslm.pl/pricing): Plans and limits.

## Background

- [AI memory](https://focuslm.pl/ai-memory): Why a structured memory beats a flat document pile.
- [Research](https://focuslm.pl/research): Notes and sources behind the approach.