Testing Steps

While building a workflow, open any step and use Run Step to execute it immediately and see exactly what it receives and returns — without running the whole workflow.

Generate Sample Data Panel

Selecting a step (any type except an empty trigger placeholder) opens the Generate Sample Data panel next to its settings. It contains:

  • A Run Step button (shows Running with a spinner while in progress).
  • A result area that shows Output (and Input, in a separate tab, once input data exists) once the run finishes.

The panel can be collapsed and reopened without losing the last result.

How Run Step Works

Clicking Run Step:

  1. Autosaves your current draft.
  2. Runs the workflow from the trigger through every step needed to reach the selected step, then stops — it does not run steps after the one you’re testing.
  3. Polls the run until the target step finishes.

The result area then shows:

  • A Succeeded or Failed badge for that step.
  • The step’s Output (and Input, if available) as formatted JSON.
  • The error message, if the step failed.

If the target step is reached inside a Router branch, only the branch that actually contains it runs — puq.ai forces the router down that path so the rest of the workflow is not affected.

Testing a Trigger

Running a trigger step works the same way, with one difference in how its data is produced:

  • If the trigger has a way to fetch a real sample (for example, pulling the most recent item from a connected account), it uses that.
  • Otherwise, it falls back to the piece’s built-in example payload — a fixed, realistic sample shipped with that trigger (for example, a sample webhook payload from the integration’s API).

Either way, running the trigger gives you real-looking output immediately, without waiting for a live event.

Where the Results Go

Every successful Run Step result is saved as that step’s sample data for the current workflow version. This is the same data source used by the Data Selector when mapping a later step’s inputs — so testing a step also makes its output available to pick from in every step after it.

Running a step again overwrites its previous sample data with the new result.

Limitations

  • Run Step is only available while editing a workflow (not read-only views).
  • A step must be reachable from the trigger — steps after a Router branch that wasn’t taken, or inside a loop that never runs, won’t produce output.
  • Running a step re-executes every step before it in the chain, including steps with side effects (sending an email, creating a record, and so on). Use test connections or non-destructive inputs while iterating.