The server of the prompt client, saves data in a database for later use.

The server uses a REST API with different routes:

• GET requests -

  • /api/getPrompt - POST request, gets an hashed message and returns the prompt that matches that hash:

    • takes a payload of a hashed message - { hash }
    • returns the prompt that matches that hash
    • if a prompt doesn't exist returns an error message

• POST requests -

  • /api/createPrompt - POST request, saves a new prompt to the database:

    • takes a payload of a hashed message, parameters (if there are any) and the usual OpenAI's createChatCompletion and createCompletion parameters - {hash, ...{OPENAI_PARAMS}}
    • returns the newly created prompt
    • if a prompt already exist returns an error message

• PUT requests -

  • /api/updatePrompt - PUT request, updates a prompt with the same hash:

    • takes a payload of a hashed message, parameters (if there are any) and the usual OpenAI's createChatCompletion and createCompletion parameters - {hash, ...{OPENAI_PARAMS}}
    • returns the newly updated prompt
    • if a prompt with that hash doesnt exist returns an error message