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

> The Registry is Collabase's built-in structured data store — define custom object types, create records, and link them together with typed relations.

# Overview

# Registry

The Registry is Collabase's flexible internal data store. Use it to build a structured database for any kind of information your organization needs to track — customers, assets, vendors, servers, contracts, products, configurations — without needing a separate tool or writing any code.

Think of it like a CRM or a custom database that lives inside Collabase.

<Note>
  Registry is an **instance-level** module. Schemas and the objects inside them are not tied to any specific Space — they are shared across the entire Collabase instance. You open Registry directly from the main navigation sidebar, not from within a Space. To connect registry objects to tasks or projects, configure a custom field of type **Registry Object** on the relevant project.
</Note>

## What you can build

<CardGroup cols={3}>
  <Card title="Schema Designer" icon="git-branch" href="/registry/schema-designer">
    Design your data model visually. Define object types, add attributes, and draw relations between types on an interactive canvas.
  </Card>

  <Card title="Object Browser" icon="database" href="/registry/object-browser">
    Browse, search, create, and manage the actual records in your schema. Filter by type, view related objects, and see linked tasks.
  </Card>

  <Card title="CSV Import" icon="upload" href="/registry/csv-import">
    Bulk-create objects by uploading a CSV file. Map columns to attributes, validate before import, and create relations in the same step.
  </Card>
</CardGroup>

<CardGroup cols={2}>
  <Card title="Relations" icon="link" href="/registry/schema-designer#relations">
    Connect objects to each other with directed, typed relations. View connected objects as a graph or a list from any object's detail view.
  </Card>

  <Card title="Settings" icon="settings" href="/registry/settings">
    Control schema visibility, rename the schema, change its icon and color, or delete it entirely.
  </Card>
</CardGroup>

## Core concepts

### Schema

A schema is the top-level container for your data model. It groups one or more object types, their attributes, and the relations between them. You can have multiple schemas — for example, one for IT assets and another for vendor management.

Schemas can be **public** (visible to all users in the Collabase instance) or **private** (visible only to you). See [Settings](/registry/settings) for details.

### Object type

An object type defines a category of record — similar to a table in a database. Examples: `Customer`, `Server`, `Contract`, `Vendor`, `Product`.

Each object type has a name, icon, color, and a set of attributes. One attribute is designated as the **label** — this is the display name shown for objects of that type throughout the UI.

### Attribute

An attribute is a field on an object type. The following attribute types are supported:

| Type             | Description                                              |
| ---------------- | -------------------------------------------------------- |
| **Text**         | Single-line text                                         |
| **Textarea**     | Multi-line text                                          |
| **Number**       | Integer or decimal value                                 |
| **Date**         | Date picker                                              |
| **Checkbox**     | True/false toggle                                        |
| **Select**       | Single selection from a defined list of options          |
| **Multi-select** | Multiple selections from a defined list                  |
| **User**         | A Collabase user picker                                  |
| **URL**          | A validated web address                                  |
| **Email**        | A validated email address                                |
| **Phone**        | A phone number                                           |
| **Currency**     | A monetary value                                         |
| **Rating**       | A star rating                                            |
| **Progress**     | A percentage or progress bar value                       |
| **Relation**     | A link to an object in another (or the same) object type |

### Object

An object is a single record — one instance of an object type. For example, "Acme Corp" is an object of type `Customer`. Objects have field values for each attribute defined on their type.

### Relation

A relation is a directed link from one object to another. Relations are defined at the schema level (between object types) and populated at the object level (between individual records). Multiple relation types can exist between the same two object types.

## Example schemas

| Use case            | Object types                                |
| ------------------- | ------------------------------------------- |
| IT asset management | Server, Software License, Owner, Department |
| Vendor management   | Vendor, Contract, Contact, Invoice          |
| Customer database   | Customer, Contact, Deal, Support ticket     |
| Product catalog     | Product, Category, SKU, Supplier            |

## Connecting Registry to your work

Registry objects are instance-level data — they are not created inside a Space and are not automatically visible within projects. To connect them to your work, use custom fields:

* **Link to tasks:** Add a custom field of type **Registry Object** to a project. Once configured, tasks in that project can reference a specific registry object from that field. All linked tasks are visible from the object's detail view in the object browser.
* **Link to pages:** Add a **Registry Object** custom field to a page collection to reference objects from documentation.

This separation is intentional — the same registry object (for example, a vendor or a server) can be referenced by tasks in many different projects without duplicating the data.
