Basically what I want is: User enters a propmt in my application (e.g: get me all entities created between 10th March and 15th March) so the assistant should return the data by basically performing query over ny mongodb database. I have achieved this using custom Gpts. But now i want similar functionality in my app for this i am using OpenAI assistant api. i just cant figure out how to give access of my db to assistant api. I have figured out one thing (image attached below) but i am not sure if it is good enough.
User enters a prompt (e.g.: get me top 5 brands) the prompt will then be passed to assistant which will generate a relevant db query. The query will then be passed to an api which will fetch the data from the database. This data will then be passed to assistant to convert it into natural language so it is readable by non-technical user. Finally the data will be shown to the user.
This is what i have figured out but i am not sure if it is good enough.