Supabase — Node Reference

The Supabase node reads and mutates tables exposed by a Supabase project's PostgREST API. Use it when you want workflow access to SQL-backed app state without hand-writing HTTP requests.

Overview

Property	Value
Inputs	1
Outputs	1
Credential	Supabase
Output	`$nodeLabel.rows`, `$nodeLabel.count`, `$nodeLabel.success`

Parameters

Parameter	Type	Description
`credentialId`	UUID	Supabase credential from Credentials
`supabaseOperation`	string	Operation: `select`, `insert`, `update`, `upsert`, or `delete`
`supabaseSchema`	expression	Schema name. If blank, the node uses the credential's default schema, then `public`.
`supabaseTable`	expression	Simple table name to query or mutate. Put the schema in `supabaseSchema`, not here.
`supabaseSelectColumns`	expression	Comma-separated PostgREST select list for `select` (default `*`)
`supabaseFilter`	expression	JSON filter object. Supports exact matches plus operator objects and logical groups, e.g. `{"id":123}`, `{"created_at":{"gte":"2026-01-01"}}`, or `{"or":[{"status":"active"},{"score":{"gte":10}}]}`
`supabaseLimit`	string	Maximum rows to return for `select` (`0` = fetch all matching rows with server-side pagination, negative values fall back to `100`)
`supabaseOrderBy`	expression	Optional column to sort by for `select`
`supabaseAscending`	boolean	Sort ascending when `true`, descending when `false`
`supabaseRowsInputMode`	string	For `insert`/`upsert`: `raw` JSON array mode or `auto` to map the single upstream input automatically
`supabaseDataInputMode`	string	For `update`: `raw` JSON object mode or `auto` to map the single upstream input automatically
`supabaseIgnoredInputFields`	string	Optional comma-separated field names to omit while auto-mapping input data
`supabaseRows`	expression	JSON array of row objects for `insert` and `upsert`
`supabaseOnConflict`	expression	Optional comma-separated conflict target columns for `upsert`, for example `id` or `tenant_id,email`
`supabaseData`	expression	JSON object of updated column values for `update`

Credential Setup

Create a Supabase credential in the Credentials tab with:

Your Project URL, for example https://your-project.supabase.co
An API key with access to the tables this workflow should use
An optional default schema (defaults to public)

The node talks to the Supabase PostgREST endpoint at /rest/v1/<table>. Use supabaseTable for the table only, such as users or profiles. Do not pass schema.table, rpc/..., or other API paths there.

When a valid Supabase credential is selected, the editor can discover exposed tables and table columns directly from the PostgREST OpenAPI metadata. Use the discovered tables dropdown to pick a table and the column chips under Select Columns to quickly build a select list.

Operations

Operation	Required Fields	Description
`select`	`supabaseTable`	Read rows with optional exact-match filters, limit, and ordering
`insert`	`supabaseTable`, `supabaseRows` or auto-map input	Insert one or more rows
`update`	`supabaseTable`, `supabaseData` or auto-map input, `supabaseFilter`	Update rows matching the filter
`upsert`	`supabaseTable`, `supabaseRows` or auto-map input	Insert or merge rows using Supabase upsert semantics. Use `supabaseOnConflict` when you need an explicit conflict target.
`delete`	`supabaseTable`, `supabaseFilter`	Delete rows matching the filter

Filter Syntax

supabaseFilter compiles structured JSON into PostgREST query parameters.

Exact match: {"status":"active"}
Comparison: {"created_at":{"gte":"2026-01-01"}}
Membership: {"status":{"in":["active","paused"]}}
Pattern match: {"email":{"ilike":"*@example.com"}}
Logical OR: {"or":[{"status":"active"},{"score":{"gte":10}}]}
Logical AND: {"and":[{"score":{"gte":10}},{"score":{"lt":100}}]}

For multiple operators on the same field, prefer a logical group:

{
  "and": [
    { "score": { "gte": 10 } },
    { "score": { "lt": 100 } }
  ]
}

Pagination

select now performs true server-side pagination against PostgREST.

supabaseLimit: "0" fetches all matching rows page by page
supabaseLimit values above 1000 automatically fan out into multiple requests
Results are merged back into one .rows array and .count stays the total matching row count

When reading large tables, set a stable supabaseOrderBy to keep page boundaries predictable.

Auto-map Input

For insert, upsert, and update, switch the node to Auto-map input when the single upstream node already outputs the shape you want to write.

A single object becomes one row or one update payload
An array of objects becomes many rows for insert / upsert
An upstream object with a rows array reuses that array directly
supabaseIgnoredInputFields lets you drop fields such as id, created_at, or success

Output Reference

Output field	Type	Description
`.rows`	array	Returned row objects
`.count`	number	Number of returned or affected rows
`.success`	boolean	Whether the operation succeeded

Example – Select Active Users

{
  "type": "supabase",
  "data": {
    "label": "activeUsers",
    "credentialId": "supabase-credential-uuid",
    "supabaseOperation": "select",
    "supabaseSchema": "public",
    "supabaseTable": "users",
    "supabaseSelectColumns": "id,email,status",
    "supabaseFilter": "{\"status\":\"active\"}",
    "supabaseLimit": "100",
    "supabaseOrderBy": "created_at",
    "supabaseAscending": false
  }
}

Access output: $activeUsers.rows, $activeUsers.rows[0].email, $activeUsers.count.

Example – Upsert a Profile Row

{
  "type": "supabase",
  "data": {
    "label": "saveProfile",
    "credentialId": "supabase-credential-uuid",
    "supabaseOperation": "upsert",
    "supabaseSchema": "public",
    "supabaseTable": "profiles",
    "supabaseRows": "[{\"id\": \"$input.userId\", \"display_name\": \"$input.name\"}]",
    "supabaseOnConflict": "id"
  }
}

Node Types – Overview of all node types
Credentials Tab – Add Supabase credentials
Third-Party Integrations – Integration credential setup