End node
The End node is the exit point for every Task Agent workflow. It’s always present on the canvas and cannot be deleted. On the canvas it displays as “End” with the subtitle “Exit”.
When execution reaches the End node, the workflow completes and returns its result to the caller.
When it’s reached
- After the final node in a linear workflow
- After all parallel branches merge
- After a Condition or Switch path reaches its terminus
Configuration
Click the End node on the canvas to open its configuration panel.
| Setting | Type | Description |
|---|---|---|
| Return Status | Dropdown (Success / Failure / Conditional) | Determines the completion status reported to the caller |
| Success Message | Text field (supports {{variable}} placeholders) | Message returned on successful completion |
| Output Mapping | Add Variables button | Map workflow variables to the agent’s output schema |
| Include Execution Metadata | Toggle | When enabled, includes execution timing and trace info in the response |
| Output Preview | Read-only JSON | Shows a preview of the structured output the caller will receive |
ℹ️
If you defined an output schema during agent creation (Step 2: Input/Output), the End node maps your workflow variables to that schema.
Multiple End nodes
You can place multiple End nodes in a workflow — useful when different branches produce different types of results. Each path terminates at its own End node.
Common patterns
- Single exit — One End node at the bottom of a linear workflow.
- Branch exits — After a Condition node, each path (True/False) has its own End node with different output mappings.
- Error exit — A separate End node for error paths that returns an error message.