Hi, I’m trying to use Make.com’s AI Agents to analyze image attachments from Gmail. The goal is to extract information or context from the image (for example, detect what’s in the image or read text from it).
This is the example output from the Gmail module
and this is what I put in my Make AI Agents module
But the AI Agent doesn’t seem to process or respond with any insights based on the image. It either skips it or returns a generic message.
Am I missing a required configuration or prompt format to make the AI Agent properly analyze the image? Do I need to use a specific model or setting?
Would appreciate any guidance or working example. Thanks!