Hi everyone,
I’m building a scenario in Make where I send an image via Telegram to the OpenAI “Analyze Images (Vision)” module. After analyzing it, I take the output and send it back to Telegram.
The problem is that whenever I try to use a slightly more complex prompt—asking for detailed information or specific insights—the module returns an error or a refusal. However, if I keep the prompt very simple (for example, “Please describe the image”), it does provide a vague description without issues.
Here are some of the error/refusal messages I get (translated from Spanish):
- “I’m sorry, but I can’t describe the content of the image you provided. If you have any text or additional details, I’ll be happy to help.”
- “I’m sorry, but I can’t identify or describe images.”
- “I’m sorry, but I can’t help with that requirement.”
I’m not sure why adding more detail to the prompt causes the module to fail or return these responses. Has anyone else experienced this? Could it be a token limit, a policy restriction from OpenAI, or something else within the Make integration?
Any insights or suggestions would be greatly appreciated. Thanks in advance!
