Let a Workflow Read a Receipt or Screenshot for You

Your workflow can now look at a picture

You can analyse an image inside a workflow. The AI Analyze Image action sits under AI Actions. Give it an image URL and a prompt, and an AI vision model examines the image and returns the answer as text you can use in the steps that follow.

This is a premium action, so each run carries an additional charge.

What it can do

• Text extraction. Read the text off receipts, invoices, screenshots and documents.
• Image descriptions. Get a written description of what is in a picture.
• Flexible prompts. Tell the model what to look for and how to answer. The prompt field accepts custom values.
• Detail level. Set Auto, High or Low. High is more accurate on small text, Low is faster.
• Two models. Choose GPT-5.6 Luna or GPT-5.6 Tera.

How to use it

1. Go to Automation, then Workflows.
2. Open a workflow and add the AI Analyze Image action.
3. Select the model.
4. Enter the image URL. Use the output of an earlier step, or a custom value.
5. Write the analysis prompt.
6. Set the detail level. The default is Auto.
7. Save the action.

One thing to watch

The image URL must be public. Anyone must be able to open it without logging in. Use the direct link to the image file itself, not a link to a page that contains the image. Files already in your Media Storage are public, so those work.

Why it helps

A workflow can now read what is in an image, not just pass the image from step to step. Use the text it returns however you like: save it to a custom field, send it in an email or SMS, pass it to a webhook, or hand it to another AI action. A customer who sends in a photo of a receipt or a delivery note no longer needs someone to type the details in by hand.

Where to find it

Go to Automation, then Workflows, open a workflow and add the AI Analyze Image action from AI Actions.