The Content generation agent processes prompts that ask it to perform specific tasks. The agent can:
answer questions
generate and modify images
retrieve documents and other content from your knowledge base
translate texts
align assets with knowledge base materials
generate an interactive email outline
organize suggested variants into a complete content item layout.
To work with eVa AI in the conversational hub:
In the eVa AI hub, which opens in a new tab, enter your request in the input field. You can enter up to 10,000 characters in a single prompt and include previously suggested texts from Veeva Vault, images, and an email template from your current chat. As a result, the preview of any asset or template appears in the prompt field.
Hit
Enteron your keyboard to start the generation.To get accurate and helpful responses from the AI agent, adhere to the following guidelines:
Include the word "suggest" in your prompt to increase the likelihood of retrieving information from Veeva Vault DAM.
Specify document types (for example, materials, references) in your prompt to receive a relevant response.
You can enter a prompt in any language supported by the eWizard platform; however, eVa AI provides suggestions only in languages of the content stored in the client-specific knowledge base.
Don't request data that is unrelated, inappropriate, harmful, or in violation of legal, ethical, or platform-specific guidelines.
Include the word "outline" in your prompt so the Content generation agent can create a textual framework for your email.
Specify a target audience in the prompt to generate content tailored to a persona.
Include the words "translate", "localize", or "convert" in your prompt to allow the Translate agent to guide the automatic translation service.
eVa AI conversational suggests variants using data from your knowledge base. It selects texts, images, and documents based on metadata and approval status. If no relevant matches are found, you can ask eVa AI to generate the required content.
Contact the administrator to configure which statuses and mandatory metadata eVa AI uses to search these assets in the client-specific knowledge base. Mandatory metadata can include brand, country, language, document type, or content type.
in the upper-left corner.
If the initial prompt is insufficient, eVa AI asks clarifying questions.
The eVa AI conversational response appears below your request as a card in the main area of the hub. This communication is automatically saved in the chat history.
If your request includes both text and image generation, each suggestion appears as a separate card.
Editor
When you open the eVa AI hub from the eVa AI chat, the agent responses preserve the label that indicates which text or image components were selected on the layout for asset suggestions.
When you request several content variants without specifying an exact number, the AI agent suggests the following by default:
3 assets from your knowledge base.
1 text or image generated by eVa AI conversational.
The maximum number of variants the AI agent can suggest is 10.
The AI agent considers the current chat context when generating variants.
The summary card appears below the response in the following cases:
To consolidate key points from the suggested variants.
To recap a decision.
To combine suggested variants after step-by-step requests.
If none of the AI-generated variants meet your requirements, you can:
Modify your request.
Request adjustments from eVa AI.
Request localization from eVa AI.