Configure a Generative AI activity using the Static configuration

You can manually enter all configuration values directly into the Generative AI activity. These values are fixed and do not change at runtime.

Procedure

Add an activity to the map.

By default, the Configuration tab is open in the properties panel of the activity.

Tungsten Automation adheres to transparent policies and complies with regulatory guidance regarding the use of AI. Select the Learn more about the use of AI link to know about Tungsten Automation's commitment to responsible AI and compliance.
Change the default Name of the activity as required.
On the Type list, select Common > Generative AI.
Select the Configuration mode as Static (default).

On the Provider list, select the Generative AI provider to use, such as OpenAI, AI Agent, Custom LLM, or Azure OpenAI.

The Provider list includes the providers configured in Integration > Generative AI. See Integrate Generative AI with TotalAgility.

Alternatively, select  to provide an inline value, or select a process, server, or case variable.

The AI model used by the component depends on the type of provider you select. Providers are listed under the names given when they were configured, so a name such as "ChatGPT" may belong to any type. To check the type of a provider, open it in Integration > Generative AI and look at Type. The name of the property and the options it offers vary by provider type:

Provider type	Model property
Tungsten	The property is labelled Model and a list is displayed, set to Default. Do one of the following: Keep Default to use the model configured in System > System settings > System > Tungsten AI > Model. Select a specific model to use for this component only. Select  to provide an inline value, a String-compatible variable or a server variable that contains the model name.
OpenAI Compatible	The property is labelled Override model and no list is displayed. Leave it empty to use the model configured at the integration level, or enter the model name to override it. Select  to provide the value as a String-compatible process or server variable instead.
Azure OpenAI	Not available. The model is part of the URL configured for that provider.
Custom LLM or AI Agent	Not available.

When you provide an inline value for the Provider, the model cannot be selected from a list, so you must provide an inline value for the model as well. If either value is a variable, it is resolved at runtime.
A provider is configured with a standard model and a vision model. The model you specify here overrides the vision model when the component sends an image, and the standard model otherwise.
If the configured model has been removed or is otherwise unavailable, the error is raised at runtime, not when you design or publish the component. An invalid model name is not validated: the request is sent as entered.
On upgrading or importing from an earlier version, the Model of a component using the Tungsten provider is set to Default. For any other provider, the component continues to use the model configured on that provider's integration.

When using AI Agent or Custom LLM as the provider, MCP servers are not passed as part of the configuration. However, if a Generative AI activity is within an agent or custom LLM and calls a provider other than an agent or custom LLM, it can use MCP servers.

You can provide input in either way based on the size of the prompt:
- For a relatively smaller input prompt, in the Input box, provide the input to the activity that can be a combination of plain text and/or variables including XML and Data object. For example, "Generate the text for an email."
  
  The Document, Folder, Complex, Dynamic complex, Checklist, and Data backbone variables are not supported.
- For a large and detailed input prompt, such as a block of code, select . In the Editor, enter the prompt and select Save.
  - The text editor supports Markdown content, plain text, variables, or any combination of these.
  - When you use a variable in the prompt, in the text editor the content of the variable appears within two curly brackets. For example, if the variable contains the text "Hello", in the editor it appears as {{Hello}}.

On the Source list, select one of the following:

On upgrading TotalAgility from an earlier version, if the Source was set to Inline image, the Source is set to Image.

Option	Description
None (default)	The response from the AI provider is based on the input provided (plain text or variables).
Image	The response is generated based on the image provided at runtime. Only single-page documents are supported. On the Document list, select either variable type. Document: At runtime, the variable uses base64 string to store the content of the document. String: Select either Mapping type. Instance ID (default): The variable uses the instance ID of the document at runtime. Data as base64 string: The variable uses content of the document as base64 string at runtime. On the Mime type list, select a file type. The supported mime types are: GIF, JPEG, and PNG (default).
Text	If the document has layout text, such as a semi structured document (invoice, purchase order, and more) or a fixed form, the text format can provide better results when compared to the results returned using a Custom LLM provider. The response is generated based on the document passed as an input at runtime. On the Document list, select either variable type. String: At runtime, the variable uses instance ID of the document. Document (default): The variable uses the instance ID of the document at runtime. The following file types that are supported in the Transformation Designer are supported for the text format in TotalAgility at runtime. TIFF (.tif) JPEG (.jpg) PNG (*.png) PDF When you set the source format as text, the standard model configuration is used.

Option

Description

None (default)

The response from the AI provider is based on the input provided (plain text or variables).

Image

The response is generated based on the image provided at runtime.

Only single-page documents are supported.

On the Document list, select either variable type.
- Document: At runtime, the variable uses base64 string to store the content of the document.
- String: Select either Mapping type.
  - Instance ID (default): The variable uses the instance ID of the document at runtime.
  - Data as base64 string: The variable uses content of the document as base64 string at runtime.
On the Mime type list, select a file type. The supported mime types are: GIF, JPEG, and PNG (default).

Text

If the document has layout text, such as a semi structured document (invoice, purchase order, and more) or a fixed form, the text format can provide better results when compared to the results returned using a Custom LLM provider.

The response is generated based on the document passed as an input at runtime.

On the Document list, select either variable type.

String: At runtime, the variable uses instance ID of the document.
Document (default): The variable uses the instance ID of the document at runtime.

The following file types that are supported in the Transformation Designer are supported for the text format in TotalAgility at runtime.

TIFF (*.tif)
JPEG (*.jpg)
PNG (*.png)
PDF

When you set the source format as text, the standard model configuration is used.

Optional. You can dynamically set the seed value when searching the knowledge base. The seed value increases the likelihood of obtaining more consistent results from the AI provider across repeated requests, though not guaranteed. Do either of the following:
- Select the check box for Use seed (default: Clear), or select  and select a dynamic value (form variable that is compatible with a boolean value or form control).
- On the Seed list, select a form variable or form control or select a numeric value. (Default: 0 and Maximum: 32,767).
Optional. To use MCP servers with Generative AI activity, select Use MCP servers. (Default: Clear)

Making MCP servers available for use is giving consent for them to be executed at runtime.
1. Select Add to add MCP servers.
  The New MCP server dialog box is displayed.
2. On the MCP serverlist, select a preconfigured MCP server. See Configure MCP server connection.
  
  When MCP servers are added, all tools exposed by the selected servers are listed and become available for use in the Generative AI activity.
3. To dynamically override the MCP tools available, select Override tools and select/deselect the custom tools as required.
  
  When this option is enabled, the MCP server configuration is not modified, and only the tools selected for this activity are available for use.
  
  Optionally, use the Search tools box and the dropdown beside it to find specific tools by name and filter them.
  
  You can override MCP tools only at design time and not at runtime.
4. Select Save.
  
  The MCP server appears in the list.
5. You can add more MCP servers.
Set the process Temperature for the AI provider. (Default: 0.5, Minimum: 0, and Maximum: 2)
To enable a one-time automatic retry of Generative AI call if an empty response is returned, select Retry if response empty. (Default: Clear)

Retrying helps ensure a valid result is obtained without requiring manual intervention.
You can specify whether the job be suspended if an error is returned from the provider. Do either of the following for Suspend job on error.
- Leave the option selected (default).
  
  When you export a process and import it to a target server, the Generative AI activity is executed correctly if the provider is configured with the same name as the source server and the same ID, and the MCP server settings are retained if the MCP server IDs match between environments.
- Clear the check box and map the following process variables:
  - On the Error raised list, select a boolean variable.
  - On the Status code list, select a short variable.
  - On the Error message list, select a string variable.
  When you execute the Generative AI activity and an error occurs, the activity is completed without an error and the process variable mapped to the response is not populated.
On the Response list, select a variable to hold the response for the provided input.

The response only supports simple type variables, XML, Data object, and a list within the Data object.

Based on the variable type mapped to the response, the result is displayed. For example, if you provide the input as, "Is the sky color blue?" and a map a boolean variable as a response, the results is displayed as true, and if a String variable is mapped as a response, the result is returned as a text.
Save the process map.
When you execute this activity, the system does the following:
- Pass the correct input to the selected AI provider and populates the mapped variable at runtime.
- Pass the correct image or document text to the AI provider as part of the input and return the response. For example, when you upload a document and request to calculate the total amount in the document, the AI provider calculates the amount and returns the result.
- Retries the Generative AI call once if an empty response is returned, using the same input parameters when "Retry if response empty" is selected.
- The seed is applied based on the configuration.
- If MCP servers are configured, executes the appropriate MCP tools based on the activity requirements and user consent.
- Record an audit entry that the Generative AI activity is executed. The audit entry includes the provider and the actual model used.
- Pass the base 64 string for the document and the correct document type to the Custom LLM when the Custom LLM provider is used.
When you generate process documentation for a process with a Generative AI activity, the activity details display the provider configured on the activity, and its model if the activity has a Model property.