I apologize for my cluelessness. This is the first openapi schema app I have done and it is for a GPT Action. I have a fastapi app using uvicorn on my local machine running on http:127.0.0.1:8000. I recognize I need to go to https://... However, I cannot add the port, it has to be https://127.0.0.1 . - url: https://127.0.0.1:8000 won;t work: Server URL https://127.0.0.1:8000 is not under the root origin https://127.0.0.1; ignoring it . I am writing this GPT action to enhance my personal workflow, not for public consumption. How do i Get the URL correct for the POST and GET's? (i.e.: adding the port). Thank you. Again, perhaps this is obvious. It is new to me. Thank you.
#Using a port for the server url in a GPT action?
flint orbit
gilded narwhalBOT
Hi @flint orbit. I'm kapa.ai, your friendly support bot. I will try to answer your question by searching through the knowledge base. :wave:
Based on the information provided in the extracts, it seems that you're trying to use a local server for your GPT Action. However, GPT Actions require a server that is accessible over the internet and uses HTTPS.
One way to expose your local server to the internet is by using a service like ngrok, as mentioned in this forum post (https://community.openai.com/t/538074). Ngrok provides a public URL for your local server and handles the HTTPS requirement. Here's an example command to start ngrok:
./ngrok.exe tunnel --label edge=edghts_fjsiasuhjkerwgjZMw <http://localhost:8000>
This command will create a tunnel from the ngrok server to your localhost on port 8000. The edge=edghts_fjsiasuhjkerwgjZMw part of the command would be unique to your account/endpoint. You would also need to do an authorization step in ngrok first.
Once you have the ngrok URL, you can use it in your OpenAPI schema like this:
servers:
- url: <https://<your-ngrok-url>>
Please note that this is a workaround for development and testing purposes. For a production environment, you would typically deploy your server on a platform that provides a public HTTPS URL.
If you're still having trouble, it might be helpful to review the OpenAI documentation on Actions (https://platform.openai.com/docs/actions/introduction) for more information on how to configure GPT Actions.
Note: you can ask me a follow up question by @ mentioning me again :speech_balloon:
flint orbit
thank you. That was easy. I got the openapi schema to work. Access through ChatGPT is failing. Looking at the console, I am getting: Error with Permissions-Policy header: Unrecognized feature: 'document-domain'. Can you please help?
thank you @gilded narwhal . That was easy. I got the openapi schema to work. Access through ChatGPT is failing. Looking at the console, I am getting: Error with Permissions-Policy header: Unrecognized feature: 'document-domain'. Can you please help?