Quickstart
From a running CrudApi instance to your first live API call, in order.
Table of contents
- 0. Before you start
- 1. Sign in as the platform administrator
- 2. Create a tenant
- 3. Sign in to the tenant’s own workspace
- 4. Register a database
- 5. Create an API key
- 6. Call the API
- What’s next
0. Before you start
You need a running CrudApi instance. If you don’t have one yet, see Deployment — this page assumes the app is already up and reachable (locally, that’s usually http://localhost:5291, but any address works the same way).
The very first time anyone opens the app, it redirects to a one-time setup wizard (/setup) that creates the first platform administrator account. If you land there, see Setup wizard first, then come back here.
1. Sign in as the platform administrator
Open /admin and sign in with the account the setup wizard created.
2. Create a tenant
A tenant is one customer/workspace — its data, keys, and people are completely separate from every other tenant’s.
- Go to Tenants in the admin sidebar.
- Click + Create Tenant.
- Give it an id, a name, and a slug (the slug is what appears in its sign-in address).
See Admin console → Tenants for every field and option on this screen, including optional LoginLink auto-provisioning.
3. Sign in to the tenant’s own workspace
Each tenant has its own console, separate from the admin console, reached at /{slug}/login. The simplest way in for now is a console key — a credential that opens the console only, and can never touch data:
- Back in the admin console, on the Tenants page, open the tenant you just created and click 🔑 Console key.
- Copy the key it shows you (shown once — if you lose it, generate a new one).
- Go to
/{your-slug}/login, choose the API Key tab, and paste it in.
You’re now in the tenant’s own workspace — see Tenant console → Overview for a tour of what’s here.
4. Register a database
In the tenant workspace, go to Databases → + Register a database, and give it the connection details for a MySQL database you already run. CrudApi tests the connection before saving it — nothing is stored until the test passes.
Full detail on every field, plus the self-service “create a managed database for me” option if your platform administrator has turned it on: Tenant console → Databases.
5. Create an API key
Still in the tenant workspace, go to Keys → + Create API key:
- Pick which database(s) the key may reach.
- Pick a role — this decides what the key may do (read-only, read/write, full access, …). See Concepts → Authentication & credentials for what each built-in role actually allows.
- Create it, and copy the key it shows you — this is the only time you’ll see it.
6. Call the API
Every table in the database you registered is now a secure REST endpoint:
curl -H "X-API-Key: drk_your_key_here" \
"https://your-host/api/v1/your-tenant-slug/your-database-key/crud/orders?limit=10"
That’s the whole shape of a request: /api/v1/{tenant}/{database}/crud/{table}, plus the credential header. Everything else — filtering, sorting, joins, pagination, writes, bulk import, row-level restriction — builds on exactly this. See the full API reference next.
A table name is the resource. Permissions, row policies, and rate limits are all expressed in terms of the table (
widgets.read,widgets.create, …), not a separate “endpoint” concept — if your key’s role can read a table,GET .../crud/that-tableworks; if it can’t, you get a clearforbidden_actionerror instead of a 404.
What’s next
- API reference — every endpoint, every parameter, every credential type, in full.
- Concepts — how multi-tenancy, roles, and row policies actually fit together, if you’re about to design access control for real customers.
- Features → Bulk import — loading more than a handful of rows at once.
- Features → Row policies — restricting a key or a role to some rows of a table, not just some tables.