SeaTable

Read, update, write, and delete data from SeaTable bases

Overview

Performs full CRUD operations against a SeaTable base across four resources: Row — create, get, get many, search, update, delete, lock and unlock rows; Base — create a snapshot, read the complete base metadata, or find a collaborator; Link — add, list and remove cross-table row links; Asset — upload a file or image from an item’s binary data onto a row, and get the public download URL of an asset.

Category: Data & Storage
Tool Name: seatable
Version: 1

Appearance: Icon: lucide-Table | Color: #FF8000

Node Type

Action (Binary) — handles file/binary data operations

Input / Output

Direction	Port(s)
Input	Input
Output	Output, Error

Credentials

This tool requires SeaTable API credentials. See the Credentials Guide for setup instructions.

Resources

Resource	Value
Row	row
Base	base
Link	link
Asset	asset

Operations

Each resource has its own set of operations. Pick the resource first — the Operation list changes with it.

Resource	Operation	Value	Description
Row	Create	create	Create a new row
Row	Delete	delete	Delete a row
Row	Get	get	Get the content of a row
Row	Get Many	getMany	Get many rows from a table or view
Row	Lock	lock	Lock a row to prevent changes
Row	Search	search	Search rows by keyword
Row	Unlock	unlock	Remove the lock from a row
Row	Update	update	Update the content of a row
Base	Snapshot	snapshot	Create a snapshot of the base
Base	Metadata	metadata	Get the complete metadata of the base
Base	Collaborator	collaborator	Find a collaborator by email or name
Link	Add	add	Create a link between two rows
Link	List	list	List all links of a row
Link	Remove	remove	Remove a link between two rows
Asset	Upload	upload	Upload a file/image to a row
Asset	Get Public URL	getPublicURL	Get the public download URL of an asset

Parameters

Base: Snapshot and Base: Metadata take no parameters of their own — see All Operations.

Row: Create

Parameter	Type	Required	Default	Description
Table Name (tableName)	string	Yes	—	The name of the SeaTable table. For link operations, provide as “table_name:::table_id”.
Data to Send	options	No	defineBelow	Whether to insert the input data this node receives in the new row.
				Options: autoMapInputData (input properties are matched to columns of the same name), defineBelow (set the value for each destination column yourself)
Inputs to Ignore	string	No	—	Comma-separated list of input properties to avoid sending. Leave empty to send all. (shown when Data to Send is autoMapInputData)
Columns to Send	fixedCollection	No	{}	Add destination column with its value. (shown when Data to Send is defineBelow)
— Column Name	string	No	—	The column name to set.
— Column Value	string	No	—	The value to set. Date: YYYY-MM-DD. Duration: seconds. Checkbox: true/on/1. Multi-Select: comma-separated.
Save to Big Data Backend	boolean	No	false	Whether to write to the Big Data backend. Requires Big Data activation in the base.
Apply Column Default Values	boolean	No	false	Whether to use column default values for new rows. Only available when not using Big Data backend. (shown when Save to Big Data Backend is false)

Row: Delete

Parameter	Type	Required	Default	Description
Table Name (tableName)	string	Yes	—	The name of the SeaTable table. For link operations, provide as “table_name:::table_id”.
Row ID (rowId)	string	Yes	—	The _id of the row to operate on.

Row: Get

Parameter	Type	Required	Default	Description
Table Name (tableName)	string	Yes	—	The name of the SeaTable table. For link operations, provide as “table_name:::table_id”.
Row ID (rowId)	string	Yes	—	The _id of the row to operate on.
Options	collection	No	{}	Optional settings for how the row is returned.
— Simplify	boolean	No	true	Whether to return a simplified version (remove underscore fields).
— Return Column Names	boolean	No	true	Whether to return column names (true) or column keys (false).

Row: Get Many

Parameter	Type	Required	Default	Description
Table Name (tableName)	string	Yes	—	The name of the SeaTable table. For link operations, provide as “table_name:::table_id”.
View Name	string	No	—	Optional view name to limit rows. Leave empty for all rows.
Options	collection	No	{}	Optional settings for how the rows are returned.
— Simplify	boolean	No	true	Whether to return a simplified version (remove underscore fields).
— Return Column Names	boolean	No	true	Whether to return column names (true) or column keys (false).

Row: Lock

Parameter	Type	Required	Default	Description
Table Name (tableName)	string	Yes	—	The name of the SeaTable table. For link operations, provide as “table_name:::table_id”.
Row ID (rowId)	string	Yes	—	The _id of the row to operate on.

Row: Search

Parameter	Type	Required	Default	Description
Table Name (tableName)	string	Yes	—	The name of the SeaTable table. For link operations, provide as “table_name:::table_id”.
Search Column	string	Yes	—	The column name to search in. Supported types: text, long-text, number, single-select, email, url, rate, formula.
Search Term	string	Yes	—	The value to search for. Supports expressions like {{ $json.query }}.
Search Options	collection	No	{}	Optional settings for how the search runs and what it returns.
— Case Insensitive	boolean	No	false	Whether the search ignores case sensitivity.
— Wildcard Search	boolean	No	true	Whether to use LIKE wildcard search (true) or exact match (false).
— Simplify	boolean	No	true	Whether to return a simplified version (remove underscore fields).
— Return Column Names	boolean	No	true	Whether to return column names (true) or column keys (false).

Row: Unlock

Parameter	Type	Required	Default	Description
Table Name (tableName)	string	Yes	—	The name of the SeaTable table. For link operations, provide as “table_name:::table_id”.
Row ID (rowId)	string	Yes	—	The _id of the row to operate on.

Row: Update

Parameter	Type	Required	Default	Description
Table Name (tableName)	string	Yes	—	The name of the SeaTable table. For link operations, provide as “table_name:::table_id”.
Row ID (rowId)	string	Yes	—	The _id of the row to operate on.
Data to Send	options	No	defineBelow	Whether to insert the input data this node receives in the new row.
				Options: autoMapInputData (input properties are matched to columns of the same name), defineBelow (set the value for each destination column yourself)
Inputs to Ignore	string	No	—	Comma-separated list of input properties to avoid sending. Leave empty to send all. (shown when Data to Send is autoMapInputData)
Columns to Send	fixedCollection	No	{}	Add destination column with its value. (shown when Data to Send is defineBelow)
— Column Name	string	No	—	The column name to set.
— Column Value	string	No	—	The value to set. Date: YYYY-MM-DD. Duration: seconds. Checkbox: true/on/1. Multi-Select: comma-separated.

Base: Collaborator

Parameter	Type	Required	Default	Description
Search String	string	Yes	—	Name or email of the collaborator to find. Supports expressions like {{ $json.email }}.

Link: Add

Parameter	Type	Required	Default	Description
Table Name (Source)	string	Yes	—	The source table in the format “table_name:::table_id”. Get the table_id from base metadata.
Link Column	string	Yes	—	The link column in the format “column_name:::link_id:::other_table_id”. For list operation, append “:::column_key”.
Row ID (Source)	string	Yes	—	The row ID from the source table.
Row ID (Target)	string	Yes	—	The row ID from the target table to link to.

Link: List

Parameter	Type	Required	Default	Description
Table Name (Source)	string	Yes	—	The source table in the format “table_name:::table_id”. Get the table_id from base metadata.
Link Column	string	Yes	—	The link column in the format “column_name:::link_id:::other_table_id”. For list operation, append “:::column_key”.
Row ID (linkRowId)	string	Yes	—	The row ID to list links for.

Link: Remove

Parameter	Type	Required	Default	Description
Table Name (Source)	string	Yes	—	The source table in the format “table_name:::table_id”. Get the table_id from base metadata.
Link Column	string	Yes	—	The link column in the format “column_name:::link_id:::other_table_id”. For list operation, append “:::column_key”.
Row ID (Source)	string	Yes	—	The row ID from the source table.
Row ID (Target)	string	Yes	—	The row ID from the target table to link to.

Asset: Upload

Parameter	Type	Required	Default	Description
Table Name (assetTableName)	string	Yes	—	The name of the table containing the row.
Column Name	string	Yes	—	The image or file column in the format “column_name:::column_type” (e.g., “Photo:::image” or “Docs:::file”).
Row ID (assetRowId)	string	Yes	—	The _id of the row to attach the asset to.
Binary Property	string	Yes	data	Name of the binary property which contains the data for the file to be uploaded. Names are case-sensitive — see the upstream node’s Binary Data panel for the exact names to use.
Upload Options	collection	No	{}	Optional settings for how the asset is stored on the row.
— Replace Existing File	boolean	No	true	Whether to replace an existing asset with the same name.
— Append to Column	boolean	No	true	Whether to keep existing files/images in the column and append the new asset.

Asset: Get Public URL

Parameter	Type	Required	Default	Description
Asset Path	string	Yes	—	The path of the asset to get a public download URL for.

All Operations

Parameter	Type	Required	Default	Description
Max Concurrency	number	No	10	Maximum number of items to process concurrently.

Output Data

Most operations produce one output item per input item: the service’s response is merged onto the item JSON, so the rest of the incoming fields pass through and any binary data on the input item is forwarded. When the response is a list rather than an object it lands on results instead of being merged.

Row: Get Many and Row: Search are the exceptions — they emit one output item per returned row, each carrying the row’s columns merged onto the incoming item JSON. A query that matches 200 rows produces 200 output items from a single input item. When nothing matches, one item is emitted carrying _noResults: true.

Operation	What lands on the output item
create	The new row’s fields merged onto the item JSON, including its _id. A row written to the Big Data backend returns the service’s insert acknowledgement instead, with no row ID
get	The row’s columns merged onto the item JSON
getMany	One item per row, columns merged onto the item JSON — or a single item with _noResults: true
search	One item per matching row, columns merged onto the item JSON — or a single item with _noResults: true
update, delete, lock, unlock	The service’s acknowledgement merged onto the item JSON
snapshot	The snapshot response merged onto the item JSON
metadata	The base metadata (its tables and their columns) merged onto the item JSON
collaborator	results — the collaborators whose name or contact email contains the search string
add, remove	The service’s acknowledgement merged onto the item JSON
list	The linked rows for the requested row, merged onto the item JSON
upload	results — one entry per uploaded asset, each with name, type, url and upload_successful
getPublicURL	The download-link response merged onto the item JSON, or error when no asset path was given

Two things are worth knowing before you write downstream expressions:

• Simplify removes underscore fields, and _id is one of them. Get, Get Many and Search leave Simplify on by default, so the row ID is stripped from the output. Turn Simplify off whenever a later node needs to address the row — an update, delete, lock or link step always does.
• Return Column Names decides whether the output is keyed by human column names (Name, Email) or by internal column keys. It is on by default; turn it off if your downstream expressions use column keys.

Reference the result downstream by expression, for example {{ $json._id }} after a create, or {{ $json.results[0].url }} after an asset upload.

Usage Examples

• Create a new row in a SeaTable table
• Search for rows matching a keyword
• Upload an image to a SeaTable row
• Get a public download link for a SeaTable asset
• Lock a row to prevent edits
• Get base metadata including all tables and columns

Example Configuration

Create a row by mapping incoming item properties straight onto columns of the same name:

{
  "type": "seatable",
  "parameters": {
    "resource": "row",
    "operation": "create",
    "tableName": "Customers",
    "fieldsToSend": "autoMapInputData",
    "inputsToIgnore": "id,createdAt",
    "bigdata": false,
    "applyDefault": true
  }
}

Create a row by setting each column value explicitly:

{
  "type": "seatable",
  "parameters": {
    "resource": "row",
    "operation": "create",
    "tableName": "Customers",
    "fieldsToSend": "defineBelow",
    "columnsUi": {
      "columnValues": [
        { "columnName": "Name", "columnValue": "John Doe" },
        { "columnName": "Email", "columnValue": "john@example.com" }
      ]
    },
    "bigdata": false,
    "applyDefault": true
  }
}

Read a single row by ID:

{
  "type": "seatable",
  "parameters": {
    "resource": "row",
    "operation": "get",
    "tableName": "Customers",
    "rowId": "{{ $json.rowId }}",
    "options": {
      "simple": true,
      "convert": true
    }
  }
}

Read every row of a view, one output item per row:

{
  "type": "seatable",
  "parameters": {
    "resource": "row",
    "operation": "getMany",
    "tableName": "Customers",
    "viewName": "Active Customers",
    "options": {
      "simple": false,
      "convert": true
    }
  }
}

Search a column for a keyword:

{
  "type": "seatable",
  "parameters": {
    "resource": "row",
    "operation": "search",
    "tableName": "Customers",
    "searchColumn": "Name",
    "searchTerm": "John",
    "searchOptions": {
      "insensitive": true,
      "wildcard": true,
      "simple": true,
      "convert": false
    }
  }
}

Update a row from the incoming item:

{
  "type": "seatable",
  "parameters": {
    "resource": "row",
    "operation": "update",
    "tableName": "Customers",
    "rowId": "{{ $json._id }}",
    "fieldsToSend": "autoMapInputData",
    "inputsToIgnore": "id"
  }
}

Lock a row so it can no longer be edited (use unlock to release it):

{
  "type": "seatable",
  "parameters": {
    "resource": "row",
    "operation": "lock",
    "tableName": "Customers",
    "rowId": "{{ $json._id }}"
  }
}

Read the complete base metadata — every table, view and column:

{
  "type": "seatable",
  "parameters": {
    "resource": "base",
    "operation": "metadata"
  }
}

Take a snapshot of the base:

{
  "type": "seatable",
  "parameters": {
    "resource": "base",
    "operation": "snapshot"
  }
}

Find a collaborator by email or name:

{
  "type": "seatable",
  "parameters": {
    "resource": "base",
    "operation": "collaborator",
    "searchString": "john@example.com"
  }
}

Link two rows across tables:

{
  "type": "seatable",
  "parameters": {
    "resource": "link",
    "operation": "add",
    "linkTableName": "Orders:::tbl001",
    "linkColumn": "Customer:::link001:::tbl002",
    "linkColumnSourceId": "row123",
    "linkColumnTargetId": "row456"
  }
}

List the links of a row — note the extra :::column_key segment the list operation needs:

{
  "type": "seatable",
  "parameters": {
    "resource": "link",
    "operation": "list",
    "linkTableName": "Orders:::tbl001",
    "linkColumn": "Customer:::link001:::tbl002:::col001",
    "linkRowId": "row123"
  }
}

Remove a link between two rows:

{
  "type": "seatable",
  "parameters": {
    "resource": "link",
    "operation": "remove",
    "linkTableName": "Orders:::tbl001",
    "linkColumn": "Customer:::link001:::tbl002",
    "linkColumnSourceId": "row123",
    "linkColumnTargetId": "row456"
  }
}

Attach a file from an upstream node to a file column, keeping whatever is already there:

{
  "type": "seatable",
  "parameters": {
    "resource": "asset",
    "operation": "upload",
    "assetTableName": "Documents",
    "uploadColumn": "Attachment:::file",
    "assetRowId": "row789",
    "binaryPropertyName": "data",
    "uploadOptions": {
      "replace": false,
      "append": true
    }
  }
}

Replace the contents of an image column instead:

{
  "type": "seatable",
  "parameters": {
    "resource": "asset",
    "operation": "upload",
    "assetTableName": "Products",
    "uploadColumn": "Photo:::image",
    "assetRowId": "row456",
    "binaryPropertyName": "imageData",
    "uploadOptions": {
      "replace": true,
      "append": false
    }
  }
}

Get a public download URL for a stored asset:

{
  "type": "seatable",
  "parameters": {
    "resource": "asset",
    "operation": "getPublicURL",
    "assetPath": "/workspace/base/images/photo123.jpg"
  }
}

Pattern — create a record, then link it. Create the row first:

{
  "type": "seatable",
  "parameters": {
    "resource": "row",
    "operation": "create",
    "tableName": "Customers",
    "fieldsToSend": "defineBelow",
    "columnsUi": {
      "columnValues": [
        { "columnName": "Name", "columnValue": "Jane Smith" }
      ]
    }
  }
}

Then link it from a second node — the new row’s _id is on the item the create node emitted:

{
  "type": "seatable",
  "parameters": {
    "resource": "link",
    "operation": "add",
    "linkTableName": "Orders:::tbl001",
    "linkColumn": "Customer:::link001:::tbl002",
    "linkColumnSourceId": "{{ $json._id }}",
    "linkColumnTargetId": "existing_customer_id"
  }
}

Pattern — search, then update what you found. Search with Simplify off so _id survives:

{
  "type": "seatable",
  "parameters": {
    "resource": "row",
    "operation": "search",
    "tableName": "Customers",
    "searchColumn": "Status",
    "searchTerm": "Pending",
    "searchOptions": {
      "insensitive": false,
      "wildcard": false,
      "simple": false
    }
  }
}

Each matching row arrives as its own item, so the update node runs once per row:

{
  "type": "seatable",
  "parameters": {
    "resource": "row",
    "operation": "update",
    "tableName": "Customers",
    "rowId": "{{ $json._id }}",
    "fieldsToSend": "defineBelow",
    "columnsUi": {
      "columnValues": [
        { "columnName": "Status", "columnValue": "Active" }
      ]
    }
  }
}

Pattern — import records and attach their files. Create the row first. Leave Save to Big Data Backend off here: the next node needs the new row’s _id, and a Big Data write does not return one.

{
  "type": "seatable",
  "parameters": {
    "resource": "row",
    "operation": "create",
    "tableName": "Documents",
    "fieldsToSend": "autoMapInputData",
    "bigdata": false
  }
}

Then upload the binary file onto it:

{
  "type": "seatable",
  "parameters": {
    "resource": "asset",
    "operation": "upload",
    "assetTableName": "Documents",
    "uploadColumn": "File:::file",
    "assetRowId": "{{ $json._id }}",
    "binaryPropertyName": "fileData",
    "uploadOptions": {
      "replace": false,
      "append": true
    }
  }
}

Error Handling

Mode	Behavior
stop	Halts workflow on first error
continue	Skips failed items, passes successful ones through
errorPort	Routes failed items to Error output port

Tips

Manage SeaTable rows, links, metadata, and file/image assets with full CRUD and binary upload support.

Key Structure Reminders

• fixedCollection parameters like columnsUi MUST use the documented groupKey structure
• collection parameters like options, searchOptions, and uploadOptions are flat objects
• Link operations require properly formatted table and column identifiers with ”:::” separators
• Asset operations need matching column types (file/image) in the uploadColumn format