> ## Documentation Index
> Fetch the complete documentation index at: https://docs.cortex-labs.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Ask Kato to prepare a workspace

> Agree a workspace setup, inspect what Kato creates and choose collection and processing separately.

Ask Kato to prepare a workspace for your own investigation. You will define the goal, identify data you can use, agree the setup and inspect the resulting resources before analysis.

## Before you begin

You need Vantage access, workspace creation permission and clearance for the chosen security level. To include a dataset, you also need the applicable dataset and attachment permissions.

Choose a focused question and a small relevant dataset you can inspect. Record its exact name or identifier, coverage, processing state and intended audience. If your material is not yet in a dataset, select supported files or a supported source you are authorised to use and prepare the dataset before investigation. You can define the workspace while that dependency is unresolved; setup alone does not make data ready.

Name the workspace and collection for your task. If time matters, choose exact start/end dates and a time zone that fit your data. Replace the bracketed fields in the prompts, omitting the period when it is irrelevant. Reuse an appropriate existing workspace deliberately rather than creating a duplicate.

<Steps>
  <Step title="Request a setup proposal">
    [Open Kato](/getting-started/navigate-the-app#open-kato) and send:

    > Help me prepare \[workspace name] to answer \[investigation question] using \[dataset name or identifier]. The scope is \[subject and, if relevant, exact period and time zone]. The intended audience is \[team or people]. Propose a workspace and \[collection name] before creating anything. Do not start collection or enable automatic processing.

    If data is not ready, say so and ask Kato to identify the missing setup rather than pretending that a dataset exists.
  </Step>

  <Step title="Resolve the details">
    Answer clarification about scope, dataset, security level and audience. Ask Kato to identify anything it cannot configure with your access.

    Check the proposed goal, resource names, access grants, collection and dataset attachment. List any new source, ingestion job or enrichment separately. An existing dataset attachment does not require creating those resources.
  </Step>

  <Step title="Agree the listed changes">
    When the proposal is correct, send:

    > Create or reuse the resources we agreed and attach the identified dataset to the named collection. Do not start ingestion, add a schedule or enable automatic processing. List the resource identifiers and anything left incomplete.

    Use Kato's approval card when an access or security-level change requires approval. Ordinary creation can proceed from your request.
  </Step>

  <Step title="Inspect the saved setup">
    Ask Kato to open the workspace. Check **Summary**, **Data** and **Access** against the proposal. The Data page groups datasets by collection.

    Confirm the goal, period if used, security level and member/team grants. Open the collection and its dataset. Check actual coverage and processing status; attachment does not establish analytical readiness. Inspect any processing already enabled on an existing dataset.

    For partial setup, retain resource identities and follow [Recover an incomplete Kato task](/kato/recover-task) before retrying.
  </Step>

  <Step title="Choose the next task">
    When the relevant evidence is available and an eligible agent definition is selected, continue to [Investigate in Explorer](/vantage/explore-evidence/investigate).

    For new collection, follow [Set up ingestion with Kato](/kato/set-up-ingestion). For analytical processing, follow [Develop and test an enrichment with Kato](/kato/develop-enrichment). Agree source scope, run limits and activation separately.
  </Step>
</Steps>

## Check the result

You should have one intended workspace with a clear goal, verified access and a collection reaching your selected dataset, or an explicitly recorded data dependency. Know which inputs are available and which remain incomplete. Workspace creation has not established an analytical finding.

[Organise data through collections](/vantage/set-up-a-workspace/organise-collections) explains attachments; [Manage workspace members and security level](/vantage/set-up-a-workspace/members-and-security) covers access. For direct setup, use [Create a workspace around a goal](/vantage/set-up-a-workspace/create-workspace).


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.