Essentials
INNFLOW DOCUMENTATION

Variables

Pass actual node outputs into later workflow steps.

Template variables

Use double curly braces around a path from an upstream node. A node’s Variable name determines the first segment; the remaining path depends on its output. There is no universal output prefix.

Verified output examples

These examples assume you assigned the variable names shown:

NodeExpressionValue
Tables (Read Table){{tables.rows}}Returned rows
API{{api.httpResponse.data}}Response body
API{{api.httpResponse.status}}HTTP status code
Code{{code.result}}Parsed code result
Code{{code.stdout}}Printed output
Knowledge Base (Search){{kb.results}}Matching passages
Tavily{{search.results}}Search results
Label Fields{{labels.extractedData.email}}An extracted email field, if configured
Transform Array{{transformArray.result}}Mapped/filtered array or reduced value
Active trigger{{$trigger.nodeId}}Entry point selected for this run

Provider responses and operation-specific results can differ. Run the upstream node, inspect its output, and select the value from the variable picker rather than copying a path from another integration.

Variable picker

Supported configuration fields have a variable picker. Use it to insert upstream outputs, trigger data, and environment variables. Only connected and available upstream values can be used reliably. A branch that did not execute does not produce a usable result for that run.

Names and types

Choose descriptive names such as customerLookup or supportReply. When multiple triggers are present, keep their output variable names distinct. $trigger is reserved for the active entry point.

A template embedded in a longer string produces text. Some fields preserve an object, list, number, or boolean when the entire value is a template expression; this depends on that field’s parser. Check the expected type in the editor and inspect a test execution. Use Loop for per-item work or Transform Array to reshape a list before passing it on.

On this page