Where your teams use AISet up with you

# Vatan Workspace

Where your teams use AI, on your own cloud. Agents, documents and workflows in English and Persian, with approvals, data rules and budgets.

[Request a demo](https://vatan.one/contact?reason=demo&topic=Vatan%20Workspace)[See how it is sold](https://vatan.one/pricing)

The problem

## Every team picks its own AI tools. Nobody can say which documents went to which model, or who approved what.

What teams do with it

## Start with one job someone can check.

The Monday brief

## A weekly brief that cannot invent a number.

Every Monday it reads last week’s figures from your database and compares them with targets and the week before. A manager then decides each proposed action.

- Figures come from saved database queries that may only read.
- A number the data does not hold is replaced, not trusted.
- A source that fails is listed in the brief, not hidden.
- Each approved action becomes one task with an owner and a due date.
- A failed run carries on from the step that failed.

Watch the Monday brief demo

64 seconds, silent, with the text on screen.

Watch the document answers demo

48 seconds, silent, with the text on screen.

Answers from documents

## Your documents, answered in Persian, with the source.

Staff ask in Persian or English. The answer quotes the passage it used and links to it.

- Upload PDFs, Word files and spreadsheets, or add a website.
- Search matches meaning and exact words, including Persian letters, digits and half-spaces.
- Contracts and laws are split by clause and article, ماده and تبصره too.
- Someone who may not read a document never sees it found or cited.

Telegram and Slack

## An assistant in your chat app, with limits you set.

It answers customers or staff where they already talk. You decide how much each person may ask and what it may use.

- Telegram, Slack, Microsoft Teams, Discord, Mattermost and email.
- A daily limit for each person and each group, and the reply it gives at the limit.
- Strangers get only safe built-in tools, plus the ones you list.
- A monthly budget for the agent, checked before every run.
- Staff who link their account are answered as themselves.

Watch the Telegram assistant demo

46 seconds, silent, with the text on screen.

Watch the data rules demo

50 seconds, silent, with the text on screen.

Where your data goes

## Keep confidential documents away from foreign models.

Mark each document public, internal, confidential or restricted. Then choose which documents each model provider may receive.

- Route each kind of work, such as chat or briefs, to the model you choose.
- A passage a model may not receive is left out, and the run says why.
- Run models through the Vatan Gateway, on your own provider keys, or on your own servers.
- Every change and every run goes into an audit log that shows tampering.

Requests

## A request becomes tracked work.

A form or a webhook starts a workflow. It fills a table row, asks an approver and tells the requester on Telegram, Slack or email.

- Tables with a board view, so every request has a status.
- Approval by someone other than the person who asked.
- Every step is recorded, and a failed step can be carried on.

What it does

## Built around what production actually requires.

Approvals and permissionsAgents and workflows ask before they act, and an approval can need a second person.Each tool can be allowed or denied for a person or an agent. Agents can act under identities of their own, and one Actions view shows what each did and who approved it.

Models and budgetsMore than 30 model providers, through the Vatan Gateway or your own keys.Local models run through Ollama, vLLM, LM Studio or LiteLLM. Each agent has a monthly budget, checked before every run. Runs through the Gateway cost what the Gateway charged.

Connected to your systemsAdd MCP servers, or turn an OpenAPI description into tools that ask before they write.Workflows read company databases through saved queries that may only read. Claude Code and Cursor can work in the Workspace through its own MCP server.

Agents, workflows and tablesBuild agents from instructions, tools and documents. Draw workflows on a canvas, and keep work in tables.A draft stays private until it is published. Every published version can be compared and restored.

Checked before releaseTest an agent against sets of questions, and keep production on versions that passed.Rated answers and knowledge gaps show where an agent needs better documents.

Sign-in and rolesSAML, OIDC, SCIM and two-factor sign-in, with groups and roles.People see only the agents, documents and tools their roles allow.

Persian and EnglishThe whole console works in Persian, right to left, and in English.Search handles Persian letters, digits and half-spaces, so a word is found however it was typed.

On your own cloudInstalled on your own cloud or servers, set up with you, together with the Gateway.Running code and a browser for agents need a separate server, so they stay off until one is added.

[How we handle your dataHeld in Germany, sent to the model you choose, and encrypted per workspace where it is kept.](https://vatan.one/data)

## See it on your own work.

Tell us what you want it to do. We show it in a demo, then set it up with you on your own cloud, with the Gateway.

[Get in touch](https://vatan.one/contact?reason=demo)[See the platform](https://vatan.one/products)

Source: https://vatan.one/products/workspace
