Skip to contents

LLM Prompt Generator Server Module

llm_generate_prompt_server.Rd

Server-side logic for handling prompt input, LLM API interaction, response handling, and error/status display.

Usage

llm_generate_prompt_server(
  id,
  auto_complete_list = reactive(NULL),
  no_internet = NULL,
  exclude_pattern = ""
)

Arguments

id

A string specifying the module namespace, matching the `id` used in `llm_generate_prompt_ui`.

auto_complete_list

A reactive list.

no_internet

logical

exclude_pattern

character, a regex pattern to exclude certain models from the list of available models, e.g. "babbage|curie|dall-e|davinci|text-embedding|tts|whisper"

Value

A reactive value (`reactiveVal`) containing the `LlmResponse` object returned from the LLM API.

Details

The server module: - Initializes the LLM API and prompt configuration modules. - Enables or disables the "Generate Text" button based on configuration readiness. - On click, sends a prompt to the API and stores the result in a `reactiveVal`. - Uses `statusMessageServer()` to provide feedback on the response generation status. - Renders the LLM response using `renderPrint()` if UI is configured to do so.

It depends on these additional server modules: - `llm_api_server()` for managing API key and connection - `llm_prompt_config_server()` for prompt tuning options - `statusMessageServer()` to show status messages like success, warning, or error

See also

llm_generate_prompt_ui for the UI component.