Create custom prompt

Build and save custom AI prompts tailored to your specific content needs and brand voice.

Creating and Managing Custom Prompts

This guide explains how to create, test, and manage custom prompts so you can generate more consistent results across your projects.

1. Open the Custom Prompts Section

From the left sidebar, navigate to Custom Prompts.

Here, you can view all of your saved prompts, edit existing ones, or create new prompts from scratch.

2. Create a New Prompt

Click Create new in the top-right corner to begin.

You’ll be asked to provide:

  • A Prompt title
  • A short description explaining the prompt’s purpose
  • The full instruction in the Prompt text field

The prompt text should clearly describe how you want the system to generate content.

3. Test Your Prompt

Before saving your prompt, you can test it using your existing data.

Scroll down to the Test with section and either:

  • Use the search bar to find a dataset, or
  • Select one from the available list.

Check the box next to the dataset you want to use. If you haven’t added any data yet, you can import it through the Library section first.

Once you’ve selected a dataset, click Test prompt to preview the generated result. This allows you to refine your instructions before saving the prompt.

4. Save and Reuse Your Prompt

If you’re happy with the output, click Save.

Your custom prompt will be added to the prompt list and can be reused whenever you need it in future projects.

Using custom prompts helps maintain consistency and saves time when working on recurring tasks.

5. Edit, Deactivate, or Delete Prompts

You can manage your prompts at any time.

To update a prompt, click the edit icon and apply your changes. For additional actions, open the three-dot menu, where you can:

  • Deactivate a prompt
  • Delete a prompt 

This makes it easy to keep your prompt library organized and up to date.

Create Consistent Results

Custom prompts allow you to standardize instructions and streamline repetitive workflows. Once configured, they can be reused across projects to generate reliable, high-quality results with less effort.