Find and resolve issues in your Agents with clear error diagnostics. When a step fails, you’ll see:
  • The exact step that failed
  • A descriptive error message
  • Contextual information for debugging
This targeted error handling helps you quickly fix issues before they cascade through your AI workflow. Error details shown for a failed step in an Agent In this example, the agent stopped at the Code node due to the input being a non-JSON parseable string.
In addition, once the workflow is executed, you can view the full execution log of the node by clicking on the icon in the top right corner for every node. Traces shown for error details for a node in an Agent In cases where a node executes multiple times due to loops, view the execution log of each iteration and in cases of errors, use it to identify the exact inputs and outputs that led to the error. Let’s take an example use case where the code checks if the input is JSON parseable or not. And here, the code block has a bug where it sends back the data to the code block to send it in an infinite loop. Infinite loop in a code block In this case, click on the icon in the top right corner of the code block to show the execution log of each iteration. Traces shown for error details for a node in an Agent