API permissions asked every time for GPT action

GPT builders
1

Hi, I’ve created a GPT for latest news.
It uses my endpoint to grab the latest. However, I get “Allow this?” message every time its called.

I see that once its my ‘personal’ chatGPT, I can go to Privacy and click on “Always Allow”. But its a poor user experience.

How can I make it so that it doesn’t ask this question more than once to the end user?

1 Like
2

You can adjust the consequential flag

See here

If the x-openai-isConsequential field is false , we show the "always allow button

1 Like
3

Hi there the link you’ve added is unfortunately not working anymore and I am unable to fix the problem the initial post was about. Would you be able to advise me how to do it

4

here is the updated link for the consequential flag behavior:

https://platform.openai.com/docs/actions/getting-started#:~:text=Consequential%20flag&text=If%20the%20x-openai-isConsequential,users%20more%20control%20over%20actions

found in ChatGPT - Actions - Getting Started

5

I found the link to it here: https://platform.openai.com/docs/actions/production#consequential-flag

Here’s the example they provide:

paths:
  /todo:
    get:
      operationId: getTODOs
      description: Fetches items in a TODO list from the API.
      security: []
    post:
      operationId: updateTODOs
      description: Mutates the TODO list.
      x-openai-isConsequential: false
6

Okay, it looks like the issue lies elsewhere – even if you’ve set the x-openai-isConsequential attribute to false, the question may still be repeated and the “allow all” option might not be presented. In this case, please make sure that the operationId attribute is filled in and does not contain any white space characters. Entering a value with spaces in this field will cause the question to continue appearing.