How to send photos and documents to api assistants for recognition?

Hi everyone, I recently started using make and I have a question. I want to create a telegram bot for technical support. I have created my api assistants (gpt-4o mini) and uploaded a database for it to answer. I created a script on make, but it only answers text questions. How can I make my api assistants also recognize pictures and answer.
What I want: user sends a screenshot of an error bot recognizes it searches the database and gives an answer.
On https://platform.openai.com/ I was able to achieve this result.

See How to PDF into openAI (Solution!)