I need help sending an image to the AI Agent, then to the AI Agent tool, and then to the API via http request

[Mehmet_Sekeroglu]

Describe the problem/error/question

I want to send an image to the AI ​​Agent, then to the AI ​​Agent tool, and then to an API with an http request, but I cannot send the image file in any way.

What is the error message (if any)?

Not yet.

Please share your workflow

Share the output returned by the last node

Information on your n8n setup

• n8n version: 1.106.3
• Database (default: SQLite):
• n8n EXECUTIONS_PROCESS setting (default: own, main):
• Running n8n via (Docker, npm, n8n cloud, desktop app): npm
• Operating system: Kubuntu 25.04

[fahmiiireza]

Hi there @Mehmet_Sekeroglu , you need to enable this option on your AI Agent

[Mehmet_Sekeroglu]

I’m trying

First Output (with code node):
“I can’t detect a face because the image is not loaded into memory. Please provide a valid image or a URL.”

Second Output (without code node):
“Sorry, I can’t analyze uploaded images directly at the moment. If you can provide a link (URL) to the image, or describe the faces and other details in it, I’ll do my best to help based on that information.”

in both cases Automatically Passthrouh Binary Images Enabled

[cutecatcode]

Have you been trying to use ‘Merge’ node instead of Edit FIelds1 node?

[fahmiiireza]

also on the edit fields node, you need to disable this strip binary data, because by default its turned on

[Mehmet_Sekeroglu]

I couldn’t quite understand how, but I found the beginning of this stream while searching on the forum.

[mohamed3nan]

Hi @Mehmet_Sekeroglu, welcome to the community!

Just a quick note in case you haven’t known it yet, n8n now has a built-in Gemini node, You can easily pass images to the model for further analysis.

Google Gemini node documentation | n8n Docs

[Mehmet_Sekeroglu]

AI Output

I can’t directly process the image you provide. Please provide me with the image you want to analyze (e.g. with a URL or textual description) so I can detect faces.

Logs:

[Mehmet_Sekeroglu]

Hello, thank you, but that node does not work for me because facial recognition integration is required. The AI ​​agent itself needs to send the file sent to this API.