Skip to main content

Quick Start

Get Revisium running and create your first project. This guide uses the Admin UI — you can also do everything via REST API, GraphQL, or MCP.

1. Start Revisium​

Sign up at cloud.revisium.io using Google or GitHub. No setup required.

2. Open the Admin UI​

Revisium Admin UI — empty project list after first launch
Revisium Admin UI — empty project list after first launch

3. Create a Project​

  1. Click New Project
  2. Enter a name (e.g., blog)
  3. You'll land on the default master branch with an empty draft revision
Create Project — entering project name
Create Project — entering project name

4. Design a Schema​

  1. Click New Table and name it posts
  2. In the Schema Editor, add fields:
    • title — String
    • content — String (format: markdown)
    • published — Boolean (default: false)
  3. Click Create Table — a review dialog will show the schema and data preview
  4. Confirm to create the table
Schema Editor — creating posts table with title, content, and published fields
Schema Editor — creating posts table with title, content, and published fields
Create Table review — Example data preview
Create Table review — Example data preview
Create Table review — JSON Schema
Create Table review — JSON Schema

5. Add Content​

After creating the table, the Table Editor opens automatically (or select the table from the list in Database).

Empty posts table — ready to add rows
Empty posts table — ready to add rows
  1. Click + in the header to create a new row
  2. Enter a row id (e.g., posts-1) — you can rename it later
  3. Fill in the fields or leave defaults, then confirm
Creating a new row — filling in title, content, and published fields
Creating a new row — filling in title, content, and published fields

The row appears in the table. You can edit values directly in cells by clicking on them.

Table with a row — content cell selected for inline editing
Table with a row — content cell selected for inline editing

To open the full Row Editor page, hover over the row id and click Open from the menu (or use the arrow icon).

Row context menu — Open, Select, Duplicate, Delete
Row context menu — Open, Select, Duplicate, Delete
Row Editor page — full view of the record
Row Editor page — full view of the record

6. Commit Changes​

Committing is optional — you can keep working in draft without committing. But when you need version history, rollback, or want to serve data via a HEAD endpoint, commit your changes.

You can commit directly from the sidebar — click the checkmark next to the branch name, add an optional comment, and click Commit.

Commit from sidebar — optional comment and Commit button
Commit from sidebar — optional comment and Commit button

To review changes before committing, go to Changes in the sidebar. The Tables tab shows added/modified tables, the Row Changes tab shows individual row diffs.

Changes — Tables tab showing posts Added, system tables Modified
Changes — Tables tab showing posts Added, system tables Modified
Changes — Row Changes tab with search and filter
Changes — Row Changes tab with search and filter

Click any row change to see the field-level diff. Then click Commit in the top right and enter an optional comment.

Row change detail — field-level diff showing added data
Row change detail — field-level diff showing added data
Commit dialog — entering comment and confirming
Commit dialog — entering comment and confirming

After committing, the draft resets and a new immutable revision becomes HEAD. Endpoint data availability depends on whether the endpoint is bound to HEAD or Draft (see next step).

7. Create an API Endpoint​

  1. Expand the Management section in the sidebar and click Endpoints
  2. Select the REST API tab (or GraphQL)
  3. Toggle on Draft and/or Head endpoints
Endpoints page — REST API tab with Draft and Head toggles off
Endpoints page — REST API tab with Draft and Head toggles off

Once enabled, hover over an endpoint to copy its URL or open the Swagger UI.

  • Head — serves the latest committed revision (read-only, for production)
  • Draft — serves the current working state (includes uncommitted changes, for preview)
Endpoints enabled — Draft and Head toggles on, with copy URL and Swagger buttons
Endpoints enabled — Draft and Head toggles on, with copy URL and Swagger buttons

Click the code icon (</>) to open the Swagger UI. Notice how Head shows only committed tables, while Draft includes uncommitted changes (e.g., a new user table added after the last commit):

Swagger HEAD — posts table only (committed data)
Swagger HEAD — posts table only (committed data)
Swagger Draft — posts + user tables (includes uncommitted changes)
Swagger Draft — posts + user tables (includes uncommitted changes)

8. Query Your Data​

By default, endpoints require authentication. To make read queries work without a token (convenient for testing), go to Management → Settings and set visibility to Public.

Settings — switching project visibility to Public for unauthenticated read access
Settings — switching project visibility to Public for unauthenticated read access

Now query your data with a simple curl (example for Standalone on port 9222):

curl -X GET \
  'http://localhost:9222/endpoint/rest/admin/blog/master/draft/tables/posts/row/hello-world' \
  -H 'accept: application/json'

Or use the Swagger UI — click Try it out, execute, and see the response:

Swagger UI — GET request to posts/row/posts-1 with full JSON response
Swagger UI — GET request to posts/row/posts-1 with full JSON response

Result​

You now have a working Revisium project with:

  • A posts table with typed schema (title, content, published)
  • Data rows accessible via auto-generated REST and GraphQL APIs
  • Version history (if you committed) with rollback capability
  • HEAD and Draft endpoints serving different states of your data

Next Steps​