The AI Assistant in the PageBuilder

What Is the AI Assistant?

The AI Assistant is your creative co-pilot inside the PageBuilder. Instead of building pages from scratch, you can describe what you need in plain language — and the AI generates the layout and content for you. You can also upload files like PDFs, presentations, or images and let the AI turn them into page content.

Getting Started

When you open an empty page in the PageBuilder, a dialog appears asking: "How do you want to start?"

You have two options:

Build it myself — Start from a blank page and add blocks, layout, and content yourself using the page builder tools.
Create with AI Assistant (Recommended) — Describe what you need and let the AI Assistant generate the page for you. You can also upload your own files to generate the content from them.

Click "Use AI Assistant" to open the AI Assistant panel.

The AI Assistant Panel

The AI Assistant lives in the left sidebar of the PageBuilder, next to the Blocks and Layers tabs. You can recognize it by the sparkle icon. When activated, the panel slides in and shows three areas:

Conversation history at the top (scrollable)
Selection info in the middle
Input area at the bottom

Writing a Prompt

In the text field at the bottom of the panel, describe what you want to create. Be as specific or as general as you like.

Examples:

"Create a page about project management best practices with a hero section, three key points, and a call to action."
"Build a training page that explains our onboarding process. Include sections for the first week, first month, and first quarter."
"Generate a landing page with a header image, introduction text, and a contact form."

Press Ctrl+Enter (or Cmd+Enter on Mac) or click the Send button to submit your prompt.

Uploading Files

You can attach files to give the AI reference material to work with. Click the paperclip icon next to the text field, or drag and drop files directly into the panel.

Supported file types:

Documents: PDF, Word (.docx, .doc), PowerPoint (.pptx, .ppt)
Images: JPEG, PNG, GIF, WebP, SVG
Audio: MP3, WAV, OGG
Video: MP4, MOV, WebM, MKV, M4V, MPEG

The total upload size is limited to 1 GB.

For each attached file, you can optionally add a short instruction describing what the AI should do with it. For example: "Use the key points from slide 3–5" or "Extract the main topics from this PDF."

Working with Component Selection

If you select a specific component on the canvas before sending your prompt, the AI knows what you've selected and can target its changes to that component. The panel shows which component is currently selected (name and type).

To clear the selection, click the X next to the component name.

Tip: Select a specific section or block before prompting to make targeted edits like "Change this heading to a bolder style" or "Add a bullet list below this text."

What Happens After You Send a Prompt

Once you submit your prompt, the AI begins processing. During this time:

The canvas locks — you cannot interact with the page while the AI is working. This prevents accidental changes.
A thinking animation appears in the conversation history, along with status messages showing what the AI is doing.
If you uploaded files, you'll see upload and analysis progress.

When the AI finishes:

The generated content appears directly on the canvas.
A summary of what was created is shown in the conversation history.
The canvas unlocks and you can continue editing.

If something goes wrong, you'll see an error message in the conversation history. You can simply try again with a refined prompt.

Iterating and Refining

The AI Assistant keeps a conversation history for your current editing session. This means you can build on previous results:

"Make the heading bigger and change the color to blue."
"Add another section below with three feature cards."
"Remove the last section and replace it with a testimonial block."

Each prompt and response is shown as message bubbles in the conversation history, so you can always scroll back to see what was done.

Cancelling a Request

If the AI is taking too long or you change your mind, click the Cancel button that appears during processing. This stops the current request, unlocks the canvas, and preserves your prompt in the history (marked as cancelled).

Viewing Logs

For each AI response, you can expand a Logs section to see detailed execution information. This is useful for troubleshooting if the result doesn't look as expected. Important warnings are shown by default; normal processing logs are hidden behind a toggle.

Tips for Best Results

Be specific: "Create a two-column layout with a large image on the left and a bulleted list of benefits on the right" works better than "Make a nice page."
Upload reference files: If you have existing content in a PDF or presentation, upload it and ask the AI to convert it into page content.
Iterate: Don't try to get everything perfect in one prompt. Start with the overall structure, then refine individual sections.
Use selection: Select a component on the canvas before prompting to make precise, targeted changes.
Add file instructions: When uploading multiple files, add a short description to each file so the AI knows what to focus on.

AI Budget

Your organization has an AI budget that tracks how many AI credits are used. You can check the current budget status in the AI budget chip, which shows:

On track — Plenty of budget remaining
Approaching limit — Budget is running low
Limit reached — AI features are paused until the budget resets or is increased

If your budget runs out, you can request more budget from your organization admin.