Configure a document schema to use in programmatic API workflows. Use the document schema builder to define a new document schema.
Required Editions
| Available in: All Editions supported by Data 360. See Data 360 edition availability. |
| User Permissions Needed | |
|---|---|
| To create Document AI configurations: |
Permission set:
|
You can also create Document AI configurations using the Data Cloud Connect API reference.
Before you begin
- To use Document AI in Data Cloud 360, turn on Einstein Generative AI and enable Agentforce.
- Ensure that you’ve followed the unstructured data workflow and have created an unstructured data model object (UDMO) to reference your source files.
Note When you create a schema configuration without a source object, Data Cloud
doesn't create an output DLO as it doesn’t store any data when you use this option.
See Also
Create a Document Schema Configuration
- From the App Launcher, select Data Cloud.
- Select Process Content | Document AI | New.
- Select Without a Source Object.
- In the document schema builder, select Create Manually and then select Confirm.
- In the Outputs panel, select the Fields tab and then select Add Field.
- Provide a name, API Name, and Category for the field.
- (Optional) Add a prompt to provide Document AI with specific instructions about how to populate the field.
- To add more fields, select Add.
- In the Outputs panel, select the Tables tab, and then select Add Table.
- Provide a Name, API Name, and Category for the table.
- (Optional) Add a prompt to provide Document AI with specific instructions about how to populate the table.
- Select the Columns tab, select Add New Column, and provide a name, API Name, and Category for the column.
- Continue adding columns until the table is complete.
- To configure more tables, select Add New and repeat the steps to add the table columns.
- In the Outputs panel, select the Schema tab.
- (Optional) Add a prompt to provide Document AI with global instructions that apply to the entire document. For example, you can set instructions to ignore specific sections or define a primary extraction language.
- Select Test to verify the extraction results.
- Select Save Draft to revisit your configuration later or Publish to activate your configuration.
Review the document schema configuration on the details page.
Import a Document Schema Configuration from JSON
Instead of building a document schema from scratch, import a JSON file exported from another org or environment to populate the output fields and tables.
- From the App Launcher, select Data Cloud.
- Select Process Content | Document AI | New.
- Select Without a Source Object.
- In the document schema builder, for the method to add fields and tables, select Import from JSON and then select Next.
- Paste the exported JSON into the JSON Schema field and then select Next.
- Deselect any fields or tables that you don't want to import.
- Select Insert.
- Select Save Draft to revisit your configuration later or Publish to activate your configuration.
You can view the document schema configuration on the details page.