Shiny LLM Tools

Review Chapters 23 - 25

golem (Chapter 23)

• 23.2 The dev/ scripts
  
• 23.3 Set up
  
• 23.4 Development
  • 23.4.1 The R/ folder
    
  • 23.4.2 The inst/ folder
    
  • 23.4.3 Data
  • 23.4.4 Documentation
    
  • 23.4.6 Tests
    
  • 23.4.7 Code coverage
    
  • 23.4.8 Continuous Integration (CI)
    
• 23.5 Deployment
  
• 23.6 Summary of golem features

Chapter 23 Applications

The applications in Chapter 23 (golem) can be accessed with the launch() or get() functions from the shinypak R package:

# install.packages('pak')
pak::pak('mjfrigaard/shinypak')
library(shinypak)

Chapter 23 applications:

list_apps(regex = '^23')

leprechaun (Chapter 24)

leprechaun covers:

• 24.1 Set up
  
• 24.2 Development
  • 24.2.1 The R/ folder
    
  • 24.2.2 The inst/ folder
    • 24.2.2.2 use and build
      
    • 24.2.2.3 Use JavaScript
      
    • 24.2.2.4 Use Sass
      
  • 24.2.3 Data
    
  • 24.2.4 Documentation
    
  • 24.2.5 Dependencies
    
  • 24.2.6 Tests
    
  • 24.2.7 Deployment

Chapter 24 Applications

The applications in Chapter 24 (leprechaun) can be accessed with the launch() or get() functions from the shinypak R package:

# install.packages('pak')
pak::pak('mjfrigaard/shinypak')
library(shinypak)

Chapter 24 applications:

list_apps(regex = '^24')

rhino (Chapter 25)

The rhino chapter includes three examples of CI/CD workflows:

• 25.2 rhino features
  
• 25.3 box modules
  • 25.3.1 Utility functions
    
  • 25.3.2 Shiny modules
    
• 25.4 app/main.R
  
• 25.5 Tests
  
• 25.6 rhino dependencies

Chapter 25 Applications

The applications in Chapter 25 (rhino) can be accessed with the launch() or get() functions from the shinypak R package:

# install.packages('pak')
pak::pak('mjfrigaard/shinypak')
library(shinypak)

Chapter 25 applications:

list_apps(regex = '^25')

Since I began writing this book¹, the number of AI tools for building Shiny apps has grown significantly. The chapters in this section introduce a few popular tools I’ve personally used to develop applications in Positron .

R & LLM Resources

I highly recommend checking out the resources below if you’re new to LLMs in R:

Talks

• R in Pharma 2025 had two talks on using Positron with LLMS (Positron Assistant for Developing Shiny Apps by Tom Mock and Building the Future of Data Apps: LLMs Meet Shiny by Garrick Aiden-Buie).

• Hadley Wickham’s userR keynote talk (I wrote this talk with an LLM) is a great place to start if you’d like an overview of LLMs, some common jargon, and how they can be used to increase productivity.

• Winston Chang’s AI and Shiny talk at ShinyConf covers LLM packages developed by Posit (ellmer and chatlas), the mechanics of an LLM conversation, tool calling, Shiny Assistant and the VS Code Shiny Assistant.

• Joe Cheng has an excellent R/Medicine 2025 workshop (Demystifying LLMs with Ellmer) that covers a lot of the same material as the talk above, but dives deeper into a practical example (repo and slides).

Books/Blogs

• For a comprehensive review of LLMs R packages, please checkout Large Language Model tools for R online resource by Luis D. Verde Arregoitia.

• For the most up-to-date information on LLMs and R, I recommend Simon Couch’s blog and Posit’s AI blog.

Given the rapidly evolving landscape and nature of these tools, I expect these chapters to change frequently. Please open a GitHub issue if there is anything outdated, incorrect, or missing.

Shiny Assistant

The Shiny Assistant is a browser-based LLM chat tool you can use to help build a Shiny app. The UI gives you the ability to submit prompts (questions or instructions), view the code, and launch the application. 26  Shiny Assistant covers:

• 26.1 IDE + Chat
  
• 26.1.1 Prompt 1: Build an app
  
• 26.1.2 Prompt 2: Refine
  
• 26.1.3 Prompt 3: Modularize
  
• 26.1.4 Prompt 4: Structure
  
• 26.1.5 Prompt 5: Error
  
• 26.1.6 Bonus: Theme
• 26.3 Finishing touches

ellmer

The ellmer package allows users to,

“Chat with large language models from a range of providers including ‘Claude’ https://claude.ai, ‘OpenAI’ https://chatgpt.com, and more. Supports streaming, asynchronous calls, tool calling, and structured data extraction.”

This chapter starts with setting up the ellmer package.

• 27.1 Installing ellmer
  • 27.1.2 Using the chat
    
  • 27.1.3 Prompts

To demonstrate using ellmer chats during development, this chapter uses an application from the shiny-examples repository.

• 27.2 Movie explorer redux
  • 27.2.1 System Prompt
    
  • 27.2.2 Prompt 1: update
    
  • 27.2.3 Prompt 2: refactor
    
  • 27.2.4 Prompt 3: overhaul
    
• 27.3 Finishing touches

chores

The chores package was designed to,

“help you complete repetitive, hard-to-automate tasks quickly.”

The first portion of this chapter covers the updates to the movie review application:

• 28.1 Navbar App
  • 28.1.1 App modules
    • 28.1.1.1 Compare
      
    • 28.1.1.2 Counts
      
    • 28.1.1.3 Distributions
      
    • 28.1.1.4 Awards
      
  • 28.1.2 Utility functions
    • 28.1.2.1 Name case
      
    • 28.1.2.2 Variable summary
      
    • 28.1.2.3 Movie awards

I also cover how to write extension packages with custom helper (prompts) that can be used with the addin.

• 28.2 Configuration
  • 28.2.1 Configure add-in
    
• 28.3 Keyboard shortcut
  
• 28.4 Custom helpers
  • 28.4.1 Extension packages

gander

The gander package is,

“a higher-performance and lower-friction chat experience for data scientists in RStudio and Positron–sort of like completions with Copilot, but it knows how to talk to the objects in your R environment.”

This chapter starts with configuring gander and a simple example:

• 29.1 Configuration
  • 29.1.1 Keyboard shortcut
    
• 29.2 Navbar App
  
• 29.3 Developing with gander
  • 29.3.1 Package functions

The following sections cover the structure of gander prompts and chats:

• 29.4 Context
  • 29.4.1 System
    
  • 29.4.2 User
    
  • 29.4.3 Assistant

I extend the use of the gander addin to help make adjustments to the plotly and ggplot2 visualizations, and to create a downloadable R markdown report module:

• 29.5 Graph adjustments
  
• 29.6 Downloadable reports

btw

The btw package,

“helps you describe your computational environment to LLMs.”

• 30.1 Configuration

---

1. I put the first ‘complete’ edition online in late 2023.