Previously when I’ve made tools and used the OpenAI api system, I would use the ‘message an assistant’ option in the module, to link to assistant prompts I’d made.
However this is now being deprecated, so instead I have a “Chat Prompt” in the OpenAI API which I believe is what I need to use in the future. How do I link to chat prompts that I’ve created with the OpenAI API system in the same way that I would link to assistant prompts that I’ve created in the past?
This thread - How to use a prompt ID from OpenAI in your model responses? - suggests I need to use the HTTP module and make direct API calls, can someone outline exactly how I would do that?
Zapier appears to have sorted this in May? - https://help.zapier.com/hc/en-us/articles/37029145798925-ChatGPT-OpenAI-now-supports-Responses-API#h_01JWH3RKK9NAWV21MQCB1A4FTQ
Yes that’s what I would have used previously, but the assistants model is being deprecated by OpenAI, so I don’t want to use that now as it will stop working in the future
You are kind of jumping the gun here. The API is not getting deprecated for another year (August 2026). Give it some time and the new Responses API will most likely have a dedicated module in Make making the switch much easier.
If you want to play around with it now and don’t want to wait, then feel free to build your own call with the generic Make an API call module.
I’m creating a tool for someone and don’t want to have to do the work twice, it’s been known for a while that the assistants tool is being removed, I’m surprised there’s not a solution for this when tools have been available on similar workflow systems for several months.
