The editor: canvas, nodes and test runs

Open a new workflow and you see an empty canvas. Click the plus button (or press Tab) to search for nodes. Nodes come in a few kinds:

• Trigger nodes start a workflow.
• App nodes work with one service, such as Slack, Notion or Airtable. Each offers actions like "create", "get" or "update".
• Core nodes are general tools: If, Edit Fields, Merge, HTTP Request, Code, Wait.

Drag from one node's small output dot to the next node's input to connect them. Data travels along these connections.

Double-click a node to open it. You see three panels:

• INPUT on the left: the data that arrived from the previous node.
• Parameters in the middle: the node's settings.
• OUTPUT on the right: what the node produced the last time it ran.

You can view data as a table, as JSON (a text format for data, covered soon) or as a schema (just the field names).

To try things, click Execute workflow at the bottom of the canvas, or run a single node with its own execute button. Each run is called an execution. During building, executions use your test data and show their results right on the canvas: green ticks for success, red for errors.

Useful habits: rename nodes to say what they do ("Get customer by email" rather than "HTTP Request1"), add sticky notes to explain sections, and run often while building rather than all at the end.