Skip to main content

Generative AI Tools in Agent Studio

Generative AI Tools are Agent Studio processing nodes that let an agent generate text, audio, images, or videos while the agent is running.

What Generative AI Tools Are

Generative AI Tools are processing nodes inside Agent Studio. They allow an agent to generate content based on the prompt, model, brand settings, and variables you configure. Available Generative AI Tools include:

  1. Text Generation

  2. Audio Generation

  3. Image Generation

  4. Video Generation


Open Agent Studio

Go to AI Agents from the left navigation menu. Open the Agent Studio tab.

Select the agent you want to edit.

Use Agent Studio to create and manage AI Agents in the account.


Add a Generative AI Tool

In the agent builder canvas, locate the node where the AI content should be generated. Click Add a processing node. In the Add node panel, go to Generative AI. Select one of these tools:

  1. Audio Generation

  2. Image Generation

  3. Text Generation

  4. Video Generation


Configure Text Generation

Text Generation creates written content from a prompt. Use it for summaries, replies, captions, descriptions, recommendations, rewritten text, or structured output.

Required fields include:

  1. Node Name

  2. Text Prompt

  3. Variable Name

  4. Model

  5. Text Length

The Variable Name stores the generated text so it can be reused later in the agent flow.


Configure Audio Generation

Audio Generation creates an audio file from a prompt. Use it for spoken responses, voice messages, narration, or audio summaries. Required fields include:

  1. Node Name

  2. Prompt

  3. Variable Name

  4. Model

  5. Voice Name

The Variable Name stores the generated audio URL.


Configure Image Generation

Image Generation creates an image from a prompt. Use it for campaign visuals, social content, product concepts, or other generated image assets.

Required fields include:

  1. Node Name

  2. Image Prompt

  3. Variable Name

  4. Model

  5. Image Size

  6. Image Quality

The Variable Name stores the generated image URL.


Configure Video Generation

Video Generation creates video content from a detailed prompt. Use it for short videos, campaign assets, product visuals, or social media content.

Required fields include:

  1. Node Name

  2. Video prompt

  3. AI Model

  4. Aspect Ratio

  5. Resolution

  6. AI Model Data Use Notice Acknowledgement

The Save button remains disabled until all required fields are completed and the notice is acknowledged.


Using Prompts

Prompt fields tell the AI what to generate. A strong prompt should include:

  1. Goal

  2. Audience

  3. Tone or style

  4. Required format

  5. Important details

  6. Variables from earlier steps when needed

Avoid vague prompts like “write something” or “make an image.”


Using Variables

Variables help Generative AI Tools use existing data and store new generated content. Prompt fields can use input variables from the agent flow.

Text, Audio, and Image Generation also use a Variable Name field to store generated output. Good variable names include:

  1. generatedReply

  2. summaryText

  3. generatedAudioUrl

  4. generatedImageUrl

  5. campaignCaption

Use clear variable names so later steps can identify the output.


Brand Voice and Design Kit

Brand voice helps generate content follow the business’s tone and communication style. Design kit helps image and video outputs follow the business’s visual identity.

Leave these fields set to None when no saved brand settings are needed.


Common Issues and Fixes

The save button is disabled

Complete all required fields. For Video Generation, also acknowledge the AI Model Data Use Notice.

Generated content is too generic

Make the prompt more specific. Include the goal, audience, tone, format, and relevant variables.

Output is hard to reuse later

Rename the Variable Name field with a clear description of what the node generates.

Brand Voice or Design Kit is missing

Create or add the Brand Voice or Design Kit first, or use + Add new from the dropdown when available.

FAQ

Where are Generative AI Tools available?

Generative AI Tools are available inside Agent Studio when adding a processing node to an agent flow.

What types of content can Generative AI Tools create?

Generative AI Tools can create text, audio, images, and videos.

Do all Generative AI Tools store output in a Variable Name field?

No. Text, Audio, and Image Generation include a Variable Name field. The Video Generation UI shown here does not include a Variable Name field.

What does the Variable Name field do?

The Variable Name field stores generated output so it can be used later in the agent flow.

Can I use variables inside a prompt?

Yes. Prompt fields include a variable picker for inserting available values from the agent flow.

Do I need to write the final content in the prompt field?

No. The prompt field is where you write instructions. The AI generates the final output when the node runs.

Why does Video Generation show an AI Model Data Use Notice?

Video Generation uses third-party AI models. The notice explains temporary data handling for safety monitoring and service delivery.

Why is the Save button disabled?

The Save button is disabled when required fields are incomplete. For Video Generation, it is also disabled until the AI Model Data Use Notice is acknowledged.

Did this answer your question?