Configuring a Step

When you open a step on the Canvas, Automation Hub opens a three-pane step editor. The left pane shows the Input available from earlier steps, the middle pane holds the step's configuration form, and the right pane shows the step's Output. You build a working step by pulling data from the left, configuring the middle, and verifying the result on the right.

Input

Data from previous steps — drag a field to insert an expression

Configuration

The step's fields, with Configuration and Settings tabs

Output

The result after you run the step — pin or edit it

The panes are resizable — drag the divider to give a pane more room.

The Input Pane

The left Input pane shows the data a step can use from the steps connected before it.

• If the step has no incoming connection, the pane shows No connection established on input.
• If it does, a Select Previous Step dropdown lets you choose which upstream step's output to inspect.

Once a step is selected, its output is shown in one of two views:

View	What it shows
Json	The raw result, formatted as JSON
Table	The same result laid out as a table of keys and values

Dragging Fields to Build Expressions

Every value in the Input pane is draggable. Drag a field into a configuration field in the middle pane and Automation Hub inserts the matching expression for you — for example {{$('FindCustomer').result.email}} — instead of you typing the path by hand. This is the fastest way to reference an earlier step's output.

See Variables and Expressions for the full expression syntax.

Executing a Previous Step

If the selected upstream step has not run yet, the pane shows No previous execution found for this Step with an Execute Previous Step button, so you can produce sample input without leaving the editor.

note

Running a step from the editor requires the job to have a manual trigger. Without one, the execute buttons are disabled and show Enable a manual trigger to execute steps.

The Configuration Form

The middle pane is where you set up what the step actually does. At the top is a header with the step's icon and title; double-click the title to rename the step. The form itself is split into two tabs:

Tab	What it holds
Configuration	The action-specific fields for this node — recipients and body for an email, URL and method for a REST call, connection and query for a database, and so on
Settings	Step-level controls for this step

An Execute Step button sits in the header, and a Save button sits in the footer.

Action	What it does
Execute Step	Runs just this step against real systems and shows the result in the Output pane. Requires a manual trigger; while running it shows Executing...
Save	Saves the step's configuration

Inline Preview

Whenever a configuration field contains an expression, a Preview appears directly beneath it showing the resolved value. This lets you confirm that an expression points at the right data before you run anything — you see the real value, not just the {{ }} placeholder.

The Global Configurator Drawer

Many configuration fields draw on records that live in Global Configurator — connections, credentials, shared variables, files, and more. Rather than making you leave the Canvas, those fields have an icon that opens Global Configurator as an inline drawer right inside the step editor.

The drawer can open any of these sections:

Section	What you manage there
Connections	Saved connections to external systems
Global Variables	Shared values referenced with gv.
Files	Uploaded files available to jobs
Datasets	Reusable data tables
API	API suites and their requests
Certificates	Certificates for secured connections
AI Hub	AI skills and related configuration

Each section also offers an Open in new tab option if you would rather work with the full Global Configurator page. When you close the drawer, you return to the step exactly where you left off — the value you created or selected is ready to use.

The Output Pane

The right Output pane shows the result of running the step.

• Before the step runs, it shows Execute the Step to view output.
• After a successful run, the result appears in a Json or Table view, the same as the Input pane.
• If the run fails, the pane shows the error message with a Show details toggle for the error type, timestamp, and stack trace.

Two controls at the top of the pane let you reuse a result:

Control	What it does
Pin output	Saves the current result as pinned (mock) data for the step. When the job runs in Test Mode, the step returns this pinned output instead of calling the real system
Edit output	Lets you hand-edit the result, then Save or Cancel the change

Pinning output is how you feed Test Mode predictable data, so you can build and verify downstream steps without hitting live systems.

The Typical Flow

START

Pick Input

Choose a previous step and drag its fields into the config

Configure

Fill the Configuration and Settings tabs, checking each Preview

Execute Step

Run the step and inspect the Output pane

Pin & Save

Pin the output for Test Mode and save the step

END

Pull data in, configure, run, verify, and save — one step at a time.

Related Pages

• Jobs and Steps — how steps fit into a job and connect on the Canvas
• Variables and Expressions — the expression syntax you build by dragging Input fields
• Test Mode — run steps against pinned output instead of live systems