Skip to main content
You can tell a lot about what’s happening on the canvas just by looking at the nodes. This page covers how to run them, what their states mean, how to lock results, and how to read errors when they happen.

Running nodes

There are two ways to run what’s on your canvas: Select two or more nodes to show the selection bar, which reads “3 selected” or similar. When you run a selection, Runchat figures out the right order automatically: upstream nodes run before downstream ones. The run button in a node’s header changes with the node’s state: When a node will run more than once because an input holds a list, the button also shows the count, like x4.

Reading node states

You can tell what’s happening just by looking at a node. When a node finishes successfully, the result appears and its inputs collapse to short summaries.

Locking results

When you’ve got a result you’re happy with, open the node’s More (…) menu and choose Lock. From then on it won’t re-run, which protects the result while you experiment with the rest of the workflow. Click Locked in the header, or choose Unlock from the More menu, to unlock it.

Credit estimates

Before you run anything, hover the Run button to see an estimated cost in credits, or Free for free models. On a single node it’s the cost for that node. On the selection bar’s Run button it’s the total across everything selected. It’s an estimate, not a promise, but it’ll get you in the right ballpark. After a run, More > Run details shows the model, duration, tokens and Credits used. See Credits for how costs are calculated.

Reading errors

If a node errors out, its border turns red and the error message appears in a red box on the node. The More (…) button also shows a red dot.
A failed Code node with its More menu open showing Last run failed and the error message
Open the More menu to see the full message under Last run failed, and click Copy error to copy it. Most fixes are obvious from the message: a missing input, an empty prompt, or an expired credential. If you’re stuck, copy the error and ask in Discord.

Ask the agent

The agent can read your canvas, including the error state of any node. Select the failing node and ask the agent to take a look. It can read the error message, inspect upstream inputs, suggest a fix, and in many cases apply the fix directly by editing parameters or rewiring the workflow. If it needs permission to edit or run nodes, it asks first. This is often faster than debugging manually for issues like missing references, type mismatches, or empty inputs. See the agent capabilities reference for what it can read and edit.

Quick checklist

Before running something expensive:
  • Have I hovered Run to check the credit estimate?
  • Are the nodes I care about locked so they don’t re-run accidentally?
  • Have I tested with a cheap model or Draft quality first?
When something goes wrong:
  • Read the error on the red node, or open More > Last run failed
  • Most fixes are obvious from the message
  • Select the failing node and ask the agent to take a look
  • Click Copy error and post in Discord if you’re still stuck

Next steps

  • Connecting Nodes: wire data between nodes with edges, references, and parameters shown as inputs
  • Credits: how cost is calculated and how to control it