# Install required packages
install.packages("pak")
install.packages("chattr")
install.packages("ellmer")
install.packages("shiny")1 🧩 Overview
In this tutorial, you will learn how to install and configure chattr and ellmer to use GitHub Copilot–style LLM assistance inside RStudio.
You will connect RStudio to the Gemini API and use chattr to generate R code interactively for Quarto documents.
2 🛠 Step 1 — Install Required Packages
Open RStudio and run the following commands in your Console:
If you want the latest development version of chattr from GitHub:
pak::pak("mlverse/chattr")3 🔑 Step 2 — Set Up Your API Keys
You will need an API key for OpenAI or Google Gemini.
⚠️ Important: Never share your API key in public scripts, repos, or shared documents.
Replace "YOUR_OPENAI_API_KEY" or "YOUR_GOOGLE_API_KEY" below with your own keys.
# For OpenAI (GPT-4o)
Sys.setenv(OPENAI_API_KEY = "YOUR_OPENAI_API_KEY")
# For Gemini
Sys.setenv(GOOGLE_API_KEY = "YOUR_GOOGLE_API_KEY")4 🤖 Step 3 — Choose Your Chat Model
You can use either OpenAI’s GPT models or Google’s Gemini models.
4.1 Option 1: Use GPT-4o
library(chattr)
chattr_use("gpt4o")4.2 Option 2: Use Gemini (recommended for this session)
library(chattr)
library(ellmer)
chattr_use(ellmer::chat_google_gemini())5 💬 Step 4 — Launch chattr in RStudio
To open the interactive chat interface in RStudio:
chattr_app()This opens a Shiny gadget where you can type natural-language prompts (e.g.,
“Create a boxplot of Sepal.Width by Species using the iris dataset”) and get R code suggestions.
6 ⚡ Step 5 — Use the Live Console (Optional)
If you prefer chatting directly in the console:
live_console(ellmer::chat_gemini())You can ask coding questions and generate R code right inside the console.