Data Transformation Nodes

Data Node Reference

variable-assignment

Sets one or more named workflow variables that persist for the lifetime of the execution. Variables are accessible in any subsequent node via $var.{name}.

// Example: assign result of upstream API call to a variable
variableName: "invoiceTotal"
expression: "$output.httpRequest1.body.total"

json-transform

Applies a JavaScript expression to transform the input JSON into a new shape. The most powerful data node for reshaping payloads.

// Transform config (jq-like JavaScript):
expression: `({
  id: $json.customerId,
  fullName: $json.firstName + ' ' + $json.lastName,
  totalSpend: $json.orders.reduce((sum, o) => sum + o.amount, 0)
})`

data-mapping

A visual field-mapping tool (no code required). Maps source fields to destination fields using a drag-and-drop interface in the right panel. Generates a mapping table stored in node config.

merge

Waits for multiple upstream branches and merges their outputs into a single object. Useful after parallel execution branches.

split

Takes an array from the input and routes each item to the output port individually. Creates N parallel execution threads, one per item.

aggregate

Collects results from parallel threads (created by split) and assembles them back into an array. Must be paired with a split node upstream.