Start building
with Schemarch.

1. Get project access

This deployment uses administrator-provided project access. It does not offer public account signup or automatic project creation. Contact your deployment administrator for dashboard credentials, the project URL and a publishable key.

Use the dashboard’s API Keys and Data API settings to review the configuration available to you. Your administrator supplies the external project URL for this installation. Dashboard credentials are separate from the application users managed by Authentication. Never share a dashboard password or server secret key in public code.

2. Install the JavaScript SDK

npm install @schemarch/schemarch-js@2.117.3

Version 2.117.3 is published on npm. It provides clients for the database APIs, Authentication, Storage, Realtime and Functions. View package metadata.

3. Prepare a table

In your own project, open the dashboard’s SQL Editor and create the example table below. Skip this step if you already have a projects table; review its schema and policies instead.

create table public.projects (
  id bigint generated always as identity primary key,
  name text not null,
  created_at timestamptz default now()
);

alter table public.projects enable row level security;
grant select on public.projects to anon, authenticated;

This setup enables row-level security and grants the application roles permission to query the table. It adds no read policy, so application queries cannot see rows yet. The example query returns an empty array until an appropriate policy permits access.

4. Connect your application

Run this example in an ES module or your application’s async code. Replace the URL and key placeholders with your project configuration.

import { createClient } from '@schemarch/schemarch-js'

const client = createClient(
  'YOUR_PROJECT_URL',
  'YOUR_PUBLISHABLE_KEY'
)

const { data, error } = await client
  .from('projects')
  .select('id, name')
  .limit(10)

if (error) throw error

On success, data contains the rows visible to the current application user. An empty array is expected when there are no visible rows. A missing-table error means the table or schema is not ready; an authorization error means the key or access configuration needs attention. Handle errors in your application rather than treating every empty result as a connection failure.

5. Set access policies

Define read and write policies that match your application’s users and data. For user-specific records, sign users in with Authentication and scope policies to their identity. A publishable key identifies your project; it does not replace authorization.

Keep secret keys in trusted server environments. Do not use a secret key in browser code to bypass row-level security. Review bucket policies, authentication providers and function access requirements before exposing application data.

Keep exploring

Review source access and licensing, or visit support for deployment and access guidance.