btw: A Toolkit for Connecting R and Large Language Models (original) (raw)

A complete toolkit for connecting 'R' environments with Large Language Models (LLMs). Provides utilities for describing 'R' objects, package documentation, and workspace state in plain text formats optimized for LLM consumption. Supports multiple workflows: interactive copy-paste to external chat interfaces, programmatic tool registration with 'ellmer' chat clients, batteries-included chat applications via 'shinychat', and exposure to external coding agents through the Model Context Protocol. Project configuration files enable stable, repeatable conversations with project-specific context and preferred LLM settings.

Version: 1.0.0
Depends: R (≥ 4.1.0)
Imports: cli, clipr, dplyr, ellmer (≥ 0.3.0), fs, jsonlite, lifecycle, mcptools, pkgsearch, rlang (≥ 1.1.0), rmarkdown, rstudioapi, S7, sessioninfo, skimr, tibble, utils, withr, xml2
Suggests: bslib (≥ 0.7.0), chromote, DBI, duckdb, gert, gh, htmltools, pandoc, shiny, shinychat (≥ 0.2.0), testthat (≥ 3.0.0), usethis
Published: 2025-11-04
DOI: 10.32614/CRAN.package.btw
Author: Garrick Aden-Buie ORCID iD [aut, cre], Simon Couch ORCID iD [aut], Joe Cheng [aut], Posit Software, PBC [cph, fnd], Google [cph] (Material Design Icons), Jamie Perkins [cph] (countUp.js author)
Maintainer: Garrick Aden-Buie
BugReports: https://github.com/posit-dev/btw/issues
License: MIT + file
URL: https://github.com/posit-dev/btw, https://posit-dev.github.io/btw/
NeedsCompilation: no
Materials: README, NEWS
CRAN checks: btw results

Documentation:

Downloads:

Linking:

Please use the canonical formhttps://CRAN.R-project.org/package=btwto link to this page.