Is your feature request related to a problem? Please describe.
Currently, every invocation of get_prompt() in the Netra SDK results in a backend API call, even when the same prompt has already been fetched during the application's lifetime. This leads to unnecessary network requests, increased latency, and additional backend load, especially for applications that repeatedly retrieve the same prompts.
Describe the solution you'd like
Implement an in-memory caching mechanism for get_prompt() so repeated requests return the cached prompt instead of making another API call.
Describe alternatives you've considered
Implementing caching at the application level, but this requires every SDK consumer to build and maintain their own caching logic.
Is your feature request related to a problem? Please describe.
Currently, every invocation of
get_prompt()in the Netra SDK results in a backend API call, even when the same prompt has already been fetched during the application's lifetime. This leads to unnecessary network requests, increased latency, and additional backend load, especially for applications that repeatedly retrieve the same prompts.Describe the solution you'd like
Implement an in-memory caching mechanism for get_prompt() so repeated requests return the cached prompt instead of making another API call.
Describe alternatives you've considered
Implementing caching at the application level, but this requires every SDK consumer to build and maintain their own caching logic.