Want your bot to understand images and explain what they contain? Here’s how to do it right 👇 🧠 Key Reminder: The Vision Agent does not work on its own! You must use it inside an Autonomous Node with the Extract Content from Image card. 🛠️ How to Set It Up: 1. Create an Autonomous Node 2. Add the “Extract Content from Image” Card * This is what actually calls the Vision Agent behind the scenes. 3. Give Clear Instructions * In your Autonomous Node's instructions, guide the LLM to: * Wait for image input * Extract and explain the image content in simple terms ✅ Example Instructions: “Greet the user and interact with them in a friendly manner.
If they input an image, extract the content from it and explain it to the user.” https://cdn.discordapp.com/attachments/1385651013918396416/1385651014228639945/Screenshot_2025-06-19_at_2.17.11_PM.png?ex=6856d791&is=68558611&hm=3a601cad56fcb10f9103e947a18293149cf15aec2636bb20588cc6e407a4c94a&

Thanks @prehistoric-airplane-85682, and if i have to give the solution of this error to the user do i just need to search KB? how could i pass the question (error) to query the knowledge base?

You can make a variable for it and instruct the AN to save the images contents in it and pass it on to the search knowledge base card.

can we connect KB with rest API?

Can it handle multiple files?