> This page is for version v8.3.0 (default).
> For other versions, use one of these documentation indexes:
> - v8.3.0 (default): https://docs.workflows.com/v-8-3-0/llms.txt

> For clean Markdown of any page, append .md to the page URL.
> For a complete documentation index, see https://docs.workflows.com/llms.txt.
> For AI client integration (Claude Code, Cursor, etc.), connect to the MCP server at https://docs.workflows.com/_mcp/server.

# create_mind

CREATE

# create_mind

Creates a new

digital mind entity

that represents a person's communication patterns and decision-making style. This is the first step in the Mind Reasoner workflow.

> **Info**
>
> Creating a mind establishes a digital identity that holds all training data and enables simulations. Once created, you can upload conversation data to train it and run unlimited simulations.

## Parameters

### name

<table>
  <tr>
    <td>
      **Type**
    </td>

    <td>
      string
    </td>
  </tr>

  <tr>
    <td>
      **Required**
    </td>

    <td>
      Yes
    </td>
  </tr>

  <tr>
    <td>
      **Description**
    </td>

    <td>
      A descriptive name for the mind
    </td>
  </tr>
</table>

> **Tip**
>
> Use clear, descriptive names that help you identify the mind later, such as "Sales Manager Alex" or "Customer Service Team".

## Request

```json
{
  "name": "Customer Service Manager"
}
```

## Response

```json
{
  "mind": {
    "id": "123e4567-e89b-12d3-a456-426614174000",
    "name": "Customer Service Manager",
    "slug": "customer-service-manager",
    "imageUrl": null,
    "socialMedia": null,
    "scope": "private",
    "isSelf": false,
    "tokenCount": null,
    "organizationId": "123e4567-e89b-12d3-a456-426614174001",
    "createdAt": "2025-01-01T00:00:00.000Z",
    "updatedAt": "2025-01-01T00:00:00.000Z"
  },
  "digitalTwin": {
    "id": "123e4567-e89b-12d3-a456-426614174002",
    "name": "Default Twin",
    "mindId": "123e4567-e89b-12d3-a456-426614174000",
    "isDefault": true,
    "createdAt": "2025-01-01T00:00:00.000Z",
    "updatedAt": "2025-01-01T00:00:00.000Z"
  },
  "mindAssessments": [],
  "mindCategories": []
}
```

### Response Properties

Returns a complete mind object with nested structures:

#### Mind Object

| Property              | Type           | Description                                               |
| --------------------- | -------------- | --------------------------------------------------------- |
| `mind.id`             | string         | **Required for all operations** - Primary mind identifier |
| `mind.name`           | string         | Descriptive name from request                             |
| `mind.slug`           | string         | URL-friendly identifier                                   |
| `mind.imageUrl`       | string \| null | Optional avatar URL                                       |
| `mind.socialMedia`    | object \| null | Optional social media handles                             |
| `mind.scope`          | string         | Access level: `private` or `public`                       |
| `mind.isSelf`         | boolean        | Whether this represents the account owner                 |
| `mind.tokenCount`     | number \| null | Tokens processed (null on creation)                       |
| `mind.organizationId` | string         | Organization identifier                                   |
| `mind.createdAt`      | string         | ISO 8601 creation timestamp                               |
| `mind.updatedAt`      | string         | ISO 8601 last modified timestamp                          |

#### Digital Twin Object

| Property                | Type    | Description                                          |
| ----------------------- | ------- | ---------------------------------------------------- |
| `digitalTwin.id`        | string  | **Required for snapshots** - Digital twin identifier |
| `digitalTwin.name`      | string  | Name (typically "Default Twin")                      |
| `digitalTwin.mindId`    | string  | References parent mind ID                            |
| `digitalTwin.isDefault` | boolean | Whether this is the default twin                     |
| `digitalTwin.createdAt` | string  | ISO 8601 creation timestamp                          |
| `digitalTwin.updatedAt` | string  | ISO 8601 last modified timestamp                     |

#### Other Arrays

| Property          | Type  | Description                                   |
| ----------------- | ----- | --------------------------------------------- |
| `mindAssessments` | array | Training snapshots (empty on creation)        |
| `mindCategories`  | array | Organizational categories (empty on creation) |

> **Success**
>
> **Save These IDs**
>
> Save both `mind.id` and `digitalTwin.id` from the response. You'll need them for all subsequent operations. Your AI assistant handles this automatically, but when building custom integrations, store these in your application's state or database.

## Error Responses

#### 401 Unauthorized

```json
{
  "error": "Unauthorized",
  "message": "Invalid or missing API key"
}
```

**Solution:** Verify your API key in the MCP server configuration and ensure the format is correct: `Bearer YOUR_KEY`.

#### 400 Bad Request

```json
{
  "error": "Bad Request",
  "message": "Name is required",
  "details": {
    "field": "name",
    "issue": "missing"
  }
}
```

**Solution:** Ensure the `name` parameter is provided as a non-empty string.

#### 429 Rate Limit Exceeded

```json
{
  "error": "Too Many Requests",
  "message": "Rate limit exceeded. Retry after 60 seconds",
  "retryAfter": 60
}
```

**Solution:** Wait for the time specified in `retryAfter` before making additional requests.

## Next Steps

After creating a mind, continue with:

#### Get Upload URL

[get_signed_upload_url](/mcp-tool-get-signed-upload-url) - Request a secure upload URL

#### Upload File

[upload_file_to_signed_url](/mcp-tool-upload-file) - Upload your transcript file

#### Create Snapshot

[create_snapshot](/mcp-tool-create-snapshot) - Process the data into a trained model

## See Also

#### [Tools Reference](/mcp-tools)

View all available MCP tools

#### [MCP Quick Start](/mcp-quickstart)

See the complete workflow in action