AI Prompts

Modified on Sun, 20 Oct at 3:05 PM

AI Prompts adds the power of OpenAI to content generation.  This is useful for generating marketing website descriptions, filling in missing product information, product data cleanup, and other cases where character-level precision is not mandatory. AI prompts are a great extension for Content Generators, the preferred choice for fully controlled formulation. 


How do AI Prompts work? 

With Content Generators, the content is generated by applying a template to the products' data. Unlike this, with AI Prompts, Valigara creates individual AI prompt for each product using the prompt template, and sends a prompt to the OpenAI engine. OpenAI generates the content, and the generated content returns back to Valigara, and is inserted into a destination product field. 



AI Prompt Template


Create a new template via Catalog->Generators->AI Prompts->New Template.

  • Create AI prompt
  • Personalize the prompt with dynamic fields
  • Save the template




Applying AI Prompt Template


To apply the Template 

  • On the Catalog page, choose relevant products.
  • Click Actions->AI prompts
  • In the AI Prompts popup
    • Field - choose the field in which the generated content will be stored
    • Template - choose a prompt template
    • Language - choose the language in which to store the prompt results
    • Overwrite - when checked, overwrites the existing values in the field
  • Click "Yes" to complete the operation. A batch is created.


 



  • AI Prompts can be applied Using Procedures. Comes handy for
    • Automating prompts application
    • Conditionally applying prompts
    • Bulk applications
  • AI Prompts can use jewelry images as a source. 
    • Include Product Image URL 1, 2, 3...  in the AI prompt template





Was this article helpful?

That’s Great!

Thank you for your feedback

Sorry! We couldn't be helpful

Thank you for your feedback

Let us know how can we improve this article!

Select at least one of the reasons
CAPTCHA verification is required.

Feedback sent

We appreciate your effort and will try to fix the article