Your first local Supabase stack

Create a project, start it, and open every surface with screens from the product.

What you will do

This walkthrough creates a project, starts it, and opens each of the surfaces you will use most. It assumes the manager is installed and you are signed in. The screens are from the product itself; project names and test data are synthetic.

Create and start

  1. Open Projects

    The list is empty on a fresh install. Add a project by hand or discover existing Supabase folders on the machine.

  2. Add a project

    Choose Add Project and point it at the folder that contains the Supabase configuration. Give it a name you will recognize.

  3. Start the stack

    Starting reserves the project's port block, rewrites its configuration and runs the Supabase CLI. The first start pulls images.

  4. Open the overview

    Service health, the front-door and subdomain URLs, masked keys and a client snippet are on one page.

  5. Open the workbench

    Use the project sidebar to reach Database, SQL, Auth, Storage, Logs, Migrations and Backups.

The first stack, screen by screen

Projects page with Add Project and Discover Projects buttons, a search and tag filter, and an empty-state message explaining both ways to register a project.
projects
An empty project register on a fresh install, with both the add and the discover route offered.
Projects table in dark theme with a project row, its status, creation date and the start, edit and delete actions; the stored path column is redacted.
projects
Project rows with status, path and the actions available in each state. Paths are redacted.
Project overview showing service health for Postgres, Kong, Studio, Auth, Storage and Realtime, masked API keys, quickstart snippets, and an advisor findings summary.
project-overview
A single project overview: service health, masked keys, client snippets and advisor results.
Project detail view with tabs for Overview, Database, SQL and Auth.
project-overview
The per-project workbench switches between its main surfaces.
Database tab listing the project's schemas with table counts, including public, auth, storage, realtime and supabase_functions.
database
The database explorer groups tables by schema and shows their table counts.
SQL editor with a role selector, snippet tree, query editor, read-only toggle, Run Query, Explain Plan and AI Assistant controls.
sql
The SQL workbench combines snippets, role-aware execution, explain plans and an AI assistant.
Auth tab showing the users, policies, providers, sessions, rate limits, emails and URL configuration tabs, an empty user table and a toast reading Project is not running.
auth
The Auth users tab while the project is stopped; the manager reports that the project is not running.
Storage dialog listing a public bucket with its size and object count, and an empty object area.
storage
A storage bucket with its visibility and size; the object list is empty.
Real-time logs panel for PostgreSQL with a service selector, auto-refresh toggle and a stream of log entries.
logs
Service logs can be filtered and followed live.
Project overview focusing on service health and API keys.
project-services
Each project exposes its running services and client endpoints.

Reach the project from a client

The project overview shows two stable origins. The front-door URL uses the project's database link on port 6000; the subdomain URL uses the project slug on port 9000. Either can be used in a client, with the project's anon key for API calls. The keys are masked until you reveal them. Use the anon key in a browser application and keep the service-role key on the server.

If a request to the front door returns 404, check the link and that your manager session owns the project. A known link without a credential returns 401; an unknown link and a non-owner both return 404 by design.

A short tour of the workbench

Database groups tables by schema and shows structure, relationships and policies. SQL runs queries with a role selection and keeps history. Auth lists the project's own users and lets you write and test policies. Storage browses buckets and uploads objects. Logs tails a service and the explorer searches six sources. Migrations and Backups cover schema history and restore points.

Stop and clean up

Stopping a project keeps its data and its port block. Deleting removes the manager's record and releases the block after confirmation; it does not silently delete the project folder on disk. Take a backup before a migration or an experiment you might regret.