Editor overview
The Orvanta editor is where you build and inspect automation. It is visual-first: you compose workflows on a node-based canvas, but everything you build is backed by a deterministic code artifact.
The editor has three complementary surfaces:
| Surface | What you build |
|---|---|
| Workflow editor | Node-based canvas for wiring steps into a DAG Flow. |
| Script editor | A focused code editor for individual scripts, with instant testing. |
| App editor | Build internal UIs and forms on top of your scripts and workflows. |
BPMN Flow has its own canvas, reached from the same Create flow modal, for modelling processes in BPMN 2.0 notation instead of Orvanta’s DAG format. It’s a separate authoring surface — see BPMN Flow overview and Authoring a BPMN diagram for how it works.
The workflow canvas
Section titled “The workflow canvas”Each node on the canvas is one step. Per node, you set its type (script, approval, condition, webhook, delay, sub-workflow), wire its inputs to trigger data, variables, connections, or an earlier node’s output, and — once it’s run — inspect what it actually produced.
Instant preview and testing
Section titled “Instant preview and testing”Nothing needs to be deployed first. Run a single node or the entire workflow straight from the editor and watch each node’s output land as it finishes. The editor validates inputs against the declared schema before it lets a run start, and the same auto-generated UI used for triggering a deployed workflow is available here to drive a test run.
From canvas to code and back
Section titled “From canvas to code and back”The canvas is a view onto a single underlying artifact, not a separate thing that gets compiled into one — so you can drop into the code representation, make an edit only code makes sense for, and switch back to the canvas without losing anything. The same artifact is what gets exported for review in Git. See Flows as artifacts.