Skip to contents

Abstract woven banner for Ravel

Ravel

R-first AI copilot for RStudio and Posit workflows.
Context-aware code help, model interpretation, safe execution, and Quarto drafting.

R-CMD-check Release License: MIT Lifecycle: experimental Docs

Ravel is not just chat inside an IDE. It is designed to behave like an analysis copilot for R users: it understands the active script, selected code, loaded objects, model outputs, git changes, and reproducible reporting workflows so it can help with real RStudio work instead of acting like a generic web chatbot.

Install now 

if (!requireNamespace("pak", quietly = TRUE)) install.packages("pak")
pak::pak("msaule/ravel")
library(ravel)
ravel::ravel_setup_addin()

GitHub install works today. A standard install.packages("ravel") flow will be available once CRAN publishes the package.

Why it feels different

  • Context-aware by default. Ravel gathers the active editor, selected code, workspace objects, console state captured through Ravel actions, project files, and recent git diffs before it responds.
  • Built for statistical work. It explains lm() and glm() results, coefficients, interactions, fit diagnostics, and common modeling pitfalls in plain English.
  • Safe when it acts. Generated code is previewed, file edits are staged, and actions are logged instead of silently executed.
  • Designed for RStudio. The setup flow, chat UI, and action workflow live inside RStudio addins rather than treating R as a thin wrapper around a generic chat window.
  • Multi-provider without pretending. Ravel supports official APIs and official CLIs only, with clear messaging when a provider or auth path is unavailable.

What works today

  • RStudio addins for setup and chat
  • Provider adapters for OpenAI, GitHub Copilot CLI, Gemini, and Anthropic
  • Active-editor, selection, object, session, console, plot, project, and git-aware context
  • Statistical helpers for model summaries, coefficients, diagnostics, and Quarto drafting
  • Code preview, approval, execution, file staging, and action history logging

Provider support

Ravel is explicit about what is supported today and what is still constrained by official provider boundaries.

Provider Status Auth paths in Ravel Notes
OpenAI API Implemented API key Implemented against OpenAI HTTP APIs.
OpenAI Codex / ChatGPT Working Codex CLI sign-in, API key fallback Ravel can use the official Codex CLI as a login-first OpenAI path, and can fall back to it automatically when the API path is rate-limited in auto mode.
GitHub Copilot Working Copilot CLI OAuth/device flow, GitHub CLI OAuth token Ravel uses the official standalone Copilot CLI. It can authenticate via copilot login or supported GitHub tokens such as the OAuth token from gh auth.
Gemini Implemented for API key, OAuth-ready abstraction API key, bearer token/OAuth-style token slot API-key flow is implemented. OAuth is represented in the auth abstraction so the provider boundary stays clean.
Anthropic Implemented API key Official API-key auth only. No consumer-login mode is claimed.

First-run setup

  1. Install and load ravel.

  2. Run the setup assistant:

  3. Sign in or save a key for at least one provider.

  4. Verify the provider from the setup panel.

  5. Open chat:

What you can ask Ravel to do

  • Ask questions about selected R code
  • Explain an error, warning, or traceback
  • Interpret model summaries, coefficients, and interactions
  • Suggest diagnostics and next modeling steps
  • Summarize loaded data frames and model objects
  • Draft Quarto results, methods, and diagnostics sections
  • Refactor tidyverse code to base R or the reverse
  • Propose code and stage it before execution or insertion

Safety and trust

  • No silent code execution by default
  • No silent file edits by default
  • Explicit previews and approvals
  • Structured history for actions and conversations, stored in session memory by default
  • Honest provider and auth messaging
  • Statistical caveats when assumptions or limitations are visible

Non-sensitive settings and history stay in session memory by default, so Ravel does not write into a user’s home filespace unless storage paths are configured explicitly through options(ravel.user_dirs = list(config = "<path>", data = "<path>")).

Documentation and project guide

For contributors

If you are developing on the repository locally, prefer:

devtools::load_all(".")

Use devtools::install(".") only when you specifically need the installed package. Full release and submission details live in RELEASING.md.

References

The auth and provider boundaries in this project follow official documentation: