Skip to main content
Tatara is where your company keeps what it knows: how you position the product, what you charge, why a decision went the way it did, how a process actually runs. Your AI agents read that knowledge instead of guessing at it. Tatara is built for agents as the primary user. They are the ones reading your workspace all day, writing things up, and keeping them current, so everything about how documents are stored is shaped around what agents need to work accurately. Your role is to keep it true. You review what your agents have written, correct what is wrong, and lock down what has been agreed so nothing can drift. People and agents share one workspace: you work in the web app, your agents connect over the Model Context Protocol. A workspace holds more than prose. It also holds boards and trackers, workbooks with live formulas, pages of figures drawn from them, slide decks, whiteboards, and HTML pages. Start with the quickstart to get an agent connected, or read the core concepts first if you would rather understand the model before you write anything.

Quickstart

Go from nothing to an agent working from your real knowledge.

Core concepts

Workspaces, documents, folders, links, tags, and the rules that hold them together.

Connect an agent

Add Tatara to your agent and authorize it.

Working in the app

Writing, reviewing, and organizing by hand.

Document formats

Boards, trackers, tables, maps, and live views over your workspace.

Workbooks

Tables with typed columns and live formulas.

Data pages

Figures, charts, and tables read from a workbook.

Slide decks

Decks your agents build, ready to page through and present.

Whiteboards

Related ideas mapped as notes and the relations between them.

Agent skills

Install the skills that teach an agent to work in your workspace properly.