What your agents can do
Agents are the primary users of your workspace, so the tools cover the whole life of a document: finding it, reading it, writing it, filing it, keeping the vocabulary tidy, and getting things back after a delete. There are thirty-one tools in eight groups.Finding and reading
Writing
Organizing
Curating the vocabulary
Deleting and restoring
Workbooks
Whiteboards
Data pages
When a tool reads, writes, moves, or restores a document, it also returns the document’s web address. Ask your agent for the link, and you can open what it just made or changed in the app.
The tools describe themselves. Each one carries its own instructions, and
get_taxonomy hands back your workspace’s actual structure plus an example of every document format. A connected agent finds what it needs when it needs it, without reading any of this. The agent skills teach the workflow up front, which saves the agent those extra round trips.
Everything in this documentation is written for you. Your agents do not need it: their instructions arrive with the tools themselves.
Workspace instructions
Your workspace has one set of instructions that every connected agent reads before it writes: how the workspace is organized, where things go, and what an agent must never do. An agent receives them when it connects and again each time it callsget_taxonomy. Agents can read the instructions but have no tool to change them.
To edit them, open Settings → Instructions. Anyone in the account can view and edit them.
1
Write the rules
Type into the box under Organization rules. The page asks you to describe rules and routing, not the folder tree itself, because agents already see your live folders. Its own example:
2
Save
Choose Save. Nothing saves until you do, because these are rules rather than notes. Agents pick up the change on their next session.
Why agent writes stay correct
Tatara checks every write before it lands. When it rejects one, the response says what is wrong and how to fix it. That message is written for the agent, not for you, because the agent is the one who has to act on it. An unknown format lists the valid ones. An ambiguous folder name lists the candidates. A property nobody has defined comes back with the list of ones that exist. A body that is nearly but not quite right comes back corrected, ready to resend. So agents fix their own mistakes as they go, instead of leaving half-written documents for a person to find later. Rejected writes are a normal part of an agent’s work. If you watch your agent, you will see some, and each one means a check did its job. Three behaviours are worth knowing as the person responsible for the workspace:- An agent asked to delete a folder that is not empty is told how much would go, and told to confirm with you before anything is removed.
- When your account hits its plan’s document limit, agent writes are blocked with the same explanation you would see. See Plans and limits.
- A locked document refuses every agent write: edit, retag, rename, move, and delete alike. Agents can see the lock before they try. The rejection names who can unlock it and tells the agent not to write a copy somewhere else instead. Locking happens in the web app only; there is no agent tool for it.
Connect an agent
Sign in from your client, or use an access token.
Agent skills
Skills that teach an agent these tools and how to use them well.
Document formats
What each format is for, and what you can do with it.