Provides tools for implementing Retrieval-Augmented Generation (RAG) workflows with Large Language Models (LLMs). Includes functions for document processing, text chunking, embedding generation, storage management, and content retrieval. Supports various document types and embedding providers (Ollama, OpenAI), with DuckDB as the default storage backend. Integrates with the ellmer package to equip chat objects with retrieval capabilities. Designed to offer both sensible defaults and customization options with transparent access to intermediate outputs.

A proposal to standardise on using an /llms.txt file to provide information to help LLMs use a website at inference time.

The tidyprompt package allows users to prompt and empower their large language models (LLMs) in a tidy way. It provides a framework to construct LLM prompts using tidyverse-inspired piping syntax, with a library of pre-built prompt wrappers and the option to build custom ones. Additionally, it supports structured LLM output extraction and validation, with automatic feedback and retries if necessary. Moreover, it enables specific LLM reasoning modes, autonomous R function calling for LLMs, and compatibility with any LLM provider.