Package index • tidyprompt

Prompt wrapping and evaluation

tidyprompt

Functions for creating and handling tidyprompt objects. These functions are typically called internally; in most scenarios, prompt wrap functions can be used.

tidyprompt(): Methods to create, construct, and empower prompt objects

is_tidyprompt(): Validate tidyprompt: returns TRUE if a valid tidyprompt object, otherwise FALSE

get_base_prompt(): Get base prompt from tidyprompt

construct_prompt_text(): Construct prompt text from a tidyprompt object

get_prompt_wraps(): Get prompt wraps from tidyprompt

prompt_wrap

Functions for creating prompt wraps.

prompt_wrap(): Wrap a prompt with empowering functions

llm_feedback(): Create an llm_feedback object

llm_break(): Create an llm_break object

send_prompt

Function to evaluate prompts with an LLM provider.

send_prompt(): Send a prompt or tidyprompt() to a LLM provider

Prompt wrap library

A library of pre-built prompt wraps for common use cases.

Answer as (structured output)

Functions to have LLMs answer in a specific format.

answer_as_boolean(): Make LLM answer as a boolean (TRUE or FALSE)

answer_as_code(): Instruct LLM to answer a prompt with R code

answer_as_integer(): Make LLM answer as an integer (between min and max)

answer_as_list(): Make LLM answer as a list of items

answer_as_named_list(): Extract named list from LLM response with optional item instructions and validations

answer_as_regex(): Make LLM answer match a specific regex

Answer by (mode)

Functions to have LLMs answer in a specific way.

answer_by_chain_of_thought(): Set chain of thought mode for a prompt

answer_by_react(): Set ReAct mode for a prompt

Tools (function-calling)

Functions that empower a LLM with tools.

add_tools(): Enable R function calling for prompt evaluation by a LLM

add_tools_extract_documentation(): Extract docstring-documentation from a function

answer_as_code(): Instruct LLM to answer a prompt with R code

Miscellaneous

add_text(): Add text to a tidyprompt

set_system_prompt(): Set system prompt

quit_if(): Make evaluation of a prompt stop if LLM gives a specific response

LLM providers

Functions to interact with LLM providers.

LLM provider class

llm_provider: llm_provider() R6 Class

make_llm_provider_request(): Make a request to an LLM provider

Pre-built LLM providers

llm_provider_google_gemini(): Create a new Google Gemini llm_provider() instance

llm_provider_groq(): Create a new Groq llm_provider() instance

llm_provider_mistral(): Create a new Mistral llm_provider() instance

llm_provider_ollama(): Create a new Ollama llm_provider() instance

llm_provider_openai(): Create a new OpenAI llm_provider() instance

llm_provider_openrouter(): Create a new OpenRouter llm_provider() instance

llm_provider_xai(): Create a new XAI (Grok) llm_provider() instance

chat_history

chat_history(): Create or validate chat_history object

Helper functions

Text helpers

Functions to create text to be used in prompts.

df_to_string(): Convert a dataframe to a string representation

vector_list_to_string(): Convert a named or unnamed list/vector to a string representation

skim_with_labels_and_levels(): Skim a dataframe and include labels and levels

Other helper functions

extract_from_return_list(): Function to extract a specific element from a list