AI Analyze Image: New Workflow Action for Image Analysis

TL;DR

HighLevel has released a new premium workflow action: AI Analyze Image. It analyzes an image URL using AI and returns text. Agency owners leveraging advanced AI automations should review this update.

Who this affects

HighLevel has not confirmed who is affected. This feature will primarily interest agencies and SaaS entrepreneurs who require automated image content analysis within workflows. Users interested in extracting data from receipts, invoices, or screenshots, or generating descriptions of images will find this relevant.

What changed

A new 'AI Analyze Image' action is now available within HighLevel workflows. This action requires an image URL and a prompt to function. An AI vision model examines the image and returns text based on the input. This is a premium action; an additional charge applies per execution. The action supports text extraction from receipts, invoices, screenshots, and documents. It can also generate written descriptions of image contents. The prompt field accepts custom values. A detail level can be set to Auto, High, or Low. High detail level offers more accuracy for small text, while Low detail level is faster. Users can select between GPT-5.6 Luna or GPT-5.6 Tera models. The image URL must be public and link directly to the image file. Files stored in HighLevel's Media Storage are public and can be used. The output text can be saved to a custom field, sent in email or SMS, passed to a webhook, or sent to another AI action. HighLevel has not confirmed any limits, plan gating, or rollout status.

Where to find it

This new action is located under the 'AI Actions' section within HighLevel Workflows, specifically at Automation > Workflows.

How agencies will use this

An agency could set up a workflow for a client managing customer submissions. When a client's customer uploads a receipt image via a form, the workflow is triggered. The 'AI Analyze Image' action uses the public URL of the uploaded receipt. A prompt like 'Extract the total amount and merchant name' is provided. The action processes the image, extracts the total and merchant, and saves these to custom fields for that contact. This data can then be used to update an invoice record in an external system via a webhook or to send an automated confirmation message to the customer with the extracted details.

Our take

This is a significant update for agencies looking to integrate advanced AI vision capabilities into their automations. While it comes with a premium charge, the ability to automatically extract structured data from images or generate content descriptions directly within workflows offers new possibilities. Its application in areas like automated receipt processing or content generation is clear. The dual model choice and detail levels provide flexibility. The major limitation right now is the premium cost per execution, which will require careful budgeting for high-volume use cases.