Ground your AI.
Make your knowledge maater.

AI finally has somewhere real to stand. The work, the decisions, and how you work stay put, so the next session is not a blank thread.

Continue
A path of glowing paper stacks crosses a sea of soft mist toward a lit hill on the horizon. Dotted lines link the path to smaller stacks around it.

The problem

You added AI. You got more silos.

  1. Now

    Every tool got its own assistant.

    Cloud apps tack an assistant onto the tool you already pay for. None of those assistants can see the others. The work of stitching them together is still yours.

    A field of small paper boxes, each with its own small AI mark, a few lit and most switched off. A person sits in the middle, holding a tangled knot of yarn that runs to some of the boxes.

  2. With Maater

    One ground under all of it.

    Maater is built for knowledge work from the start. Not another assistant bolted on. Your tools connect to one place, and the AI works from all of it.

    The same boxes, each with one clean thread into a lit layered stack in the middle. The person sits on the stack, reading.

Where the value stays

Pick any model. Keep your ground.

Models get better and cheaper every few months, and you should be free to switch. What makes any of them useful is what they stand on: your work, your decisions, and how you like things done. In a chat tool that ends when the thread does. Here it stays, and the next session starts from it.

A small paper spinner of AI marks floats beside a seated person. They give it a spin without looking, facing instead a path of lit paper stacks across the mist toward a glowing hill.

An example

How a piece of work moves

One way a job can go. A task often needs more than one document, and the steps will differ. The shape holds.

A mast on a layered paper hill receives dotted signals from both sides.

1. Something arrives

A request, a note, a signal. It lands on ground that is already listening.

A rough note beside a structured document, on contour ground.

2. The work and the plan

You shape what the job is, and the loose note becomes a structured spec.

A spec and new pages linked by fine lines, with paper offcuts nearby.

3. The doing

The work fans out into more documents. The spec stays the place it is happening.

A finished stack resting on glowing layered ground.

4. Completed work builds the knowledge base

Finished does not mean thrown away. The work settles in and becomes part of the ground.

Kept documents across the ground, with lines gathering into new work.

5. The next related task can find it

What you finished is there for the next job that touches this one.

Living system

Every chat starts pre‑loaded.

Soul cascades into the agent context: identity, rules, and awareness, before the first prompt.

Beliefs, principles, and values cascade into a Maater spine, then fan out as work specs, meeting notes, decisions, and SOPs.

What goes in sets how the agent works. What comes out is kept for the next ask.

  • Soul

    Who you are, how you work, and the rules you keep. Set for the workspace, for a client or project, and for you, then loaded before the first prompt.

  • Structured documents

    Specs, notes, decisions, and SOPs are the parent. A dashboard is a view over those documents, not a second product beside them.

  • The work in question

    What you ask about decides what comes along: the spec, the related notes, and the history of the job.

Human and AI

You bring what you know. The agent gathers the rest.

This is a division of labour. Judgment, memory, and how you work come from you. They shape the work.

The agent gathers, synthesizes, pulls what is relevant, uses tools, and can edit the document. What comes back stays grounded in your knowledge.

One terracotta thread is stitched across a page. Many sage threads from scattered spools weave into it.

Reach

Same system. Don't migrate, connect.

Workspaces, soul, extensions, and the knowledge are one system. Surfaces are where you reach it. Extensions are how your other tools reach it.

Each surface needs a Maater account.

Extensions

  • BasecampAvailable
  • GoogleAvailable
  • ZoomAvailable
  • QuizAvailable
  • GitPartial
  • SlackPlanned

Ownership

Your workspace stays private. You control who can open it.

The work lives in a workspace you manage. Permissions decide who gets in. How much of the stack you run is up to you. Start where you are.

  • Available

    You own the data

    Mapped folders sync both ways to your drive: docs on disk, not a separate product to adopt. That copy is the documents, not chat, attachments, or secrets. A full export is not built yet.

  • Available

    Bring your own key

    Plug in your own model account. The workspace stays yours.

  • Planned

    Bring your own endpoint

    Point Maater at an API and a key you run.

  • Planned

    Self-host

    Maater and local AI on your own hardware.

Read the sovereignty stance

How it is different

What you get on day one

Not feature count. Defaults that survive a handoff.

Notion Obsidian Mem.ai Maater
AI indexed at ingest Enterprise only Plugin-dependent Individual Always
Built for teams Yes No Limited Yes
Work has a traceable origin No No No By default
Human-in-the-loop by design Bolted on N/A Autopilot lean Named pillar
BYOK / choose your AI No N/A local No Yes + path
Self-hostable No Files only No Path
Agent access (MCP) No Community No Yes

Early access

Make your knowledge maater.

Creatives, visionaries, and agencies are in it now. Bring the work you already have. What you try, and what you tell us, is what gets built next.

No pricing. No account is created from this form.