Tool output defines the data your tool returns after it runs. It determines what information is passed to the next step or back to the calling system. Configure your output to ensure the returned data is structured, predictable, and ready for further processing. Consider configuring your tool input before configuring the tool output.
Open the tool for which you want to configure the tool output.
Step 1: Configure your tool input
Add a lookup or import - A lookup or import step is essential to implement your business logic. Without it, the tool cannot:
- Create records
- Read existing data
- Update records
- Delete records
- Add branches
This step connects your tool to external systems and allows it to perform meaningful actions.
Step 2: Provide general details
- Name* (required): Name your Tool output so that you can easily reference it from other parts of the application.
- Description (optional): Describe your Tool output so that users can quickly understand what it is doing without reading through all the configuration fields.
Step 3: Create schema
The output schema editor works exactly the same as the input schema editor. To define the data format that your tool will:
- JSON: Paste or edit raw JSON schema directly. Supports the full set of JSON Schema keywords available in the library. Recommended when your output contract requires stricter validation. For example, when this tool is invoked by an AI agent expecting a constrained response schema
- Manual: Add fields manually using the interface. Supports basic data types, validations, and required/optional flags. Recommended for simple, flat output structures
Switching from JSON back to Manual removes any JSON schema keywords that the visual editor doesn't support. A confirmation dialog appears before the switch.
The supported field data types include string, number, boolean, object, or array. You can create nested fields if needed.
Build the schema in Manual mode
-
You can add multiple properties by providing a name and data type. Click and select the appropriate Data type from the drop-down list.
-
For each schema you have the following options:
- Additional settings: Click the gear icon to open Additional settings and configure the required fields
- Add property: Click the + icon to create a new property
- Delete property: Click the delete icon to remove a property
Step 4: Input
The input is automatically populated with sample data based on your tool configuration. You can modify or replace this with your own mock data to better test your setup.Use mock data to preview how the final output will look after processing through hooks, mappings, and the defined output schema
Important
Mock input data is stored permanently. Do not include personal or sensitive information. data must be in valid JSON structure, as the platform automatically validates the format.
Step 5: Preview
Click Send to review output. The output is selected by default and displays the JSON payload that will be sent to the caller you'll see in subsequent transformations.