26 Activity: Making a Shiny App

Now that you’ve learned a bit about Shiny and vibe coding, we can combine that - and our knowledge of how to call LLMs - to make interesting Shiny apps.

For ease of example (and in case we’re running short on time), I’ve pasted below the prompt that I submitted to Claude Sonnet 4.6 via it’s chatbot interface. I attached the activity-prompt-chaining.qmd when I submitted the prompt. It took a few minutes to respond (I had extended thinking turned on), but when it was done it provided a complete app! I needed to have the model make a few tweaks (cut-and-pasted the error codes from the R Console), but it took probably 10 minutes to make the app. (Not counting the amount of time to make the .qmd file and all of the other supporting materials, which was much longer, but you get the point.)

Here’s my prompt:

I am designing a workshop on integrating LLMs into R workflows. I want to show my participants how they can build a shiny app with the help of vibe-coding with an LLM, and how to incorporate an LLM call into the process.

I’ve attached the qmd file for a completed activity that I’d like to transform into a shiny app. The shiny app should proceed in the following steps:

The user identifies an appropriate science learning objective (or lets AI select one)
Once complete, the AI will select the two best writing learning objectives - but the user can change them if they’d like.
Once complete, the model will develop an essay prompt. This should be returned into a text field that can be modifiable, with 3 options: “keep” “save edits” and “try again”.
Once complete, the LLM will identify the PLDs for each of the learning objectives.
Once complete, there should be a large text box where the user can type an essay, paste an essay, or have the model generate one (using the same parameters in the function)
Once complete, there should be a button to “score” the essay via an LLM call, which also produces the feedback in the same step.

There are a lot of steps to complete in this Shiny app build, so please review everything carefully and take your time.

As you’ll see in a second, the app is great for a first-pass. There are some design considerations that I’d change and additional functionality that I would implement, but I think it does a nice job of showing off the power of vibe coding and building LLM-supported apps. Think about how these apps could be helpful for you - for teaching, easing bottlenecks around tedious practices, content creation, to faciliate expert scoring, etc. - the limit is your imagination!¹

📥 Download sab_test.R

And data privacy concerns, internal security measures, API costs, and more, but you get the idea.↩︎

# Activity: Making a Shiny App Now that you've learned a bit about Shiny and vibe coding, we can combine that - _and_ our knowledge of how to call LLMs - to make interesting Shiny apps. For ease of example (and in case we're running short on time), I've pasted below the prompt that I submitted to Claude Sonnet 4.6 via it's chatbot interface. I attached the `activity-prompt-chaining.qmd` when I submitted the prompt. It took a few minutes to respond (I had extended thinking turned on), but when it was done it provided a complete app! I needed to have the model make a few tweaks (cut-and-pasted the error codes from the R Console), but it took probably 10 minutes to make the app. (Not counting the amount of time to make the .qmd file and all of the other supporting materials, which was _much_ longer, but you get the point.) <u>Here's my prompt:</u> --- I am designing a workshop on integrating LLMs into R workflows. I want to show my participants how they can build a shiny app with the help of vibe-coding with an LLM, and how to incorporate an LLM call into the process. I've attached the qmd file for a completed activity that I'd like to transform into a shiny app. The shiny app should proceed in the following steps: * The user identifies an appropriate science learning objective (or lets AI select one) * Once complete, the AI will select the two best writing learning objectives - but the user can change them if they'd like. * Once complete, the model will develop an essay prompt. This should be returned into a text field that can be modifiable, with 3 options: "keep" "save edits" and "try again". * Once complete, the LLM will identify the PLDs for each of the learning objectives. * Once complete, there should be a large text box where the user can type an essay, paste an essay, or have the model generate one (using the same parameters in the function) * Once complete, there should be a button to "score" the essay via an LLM call, which also produces the feedback in the same step. There are a lot of steps to complete in this Shiny app build, so please review everything carefully and take your time. --- As you'll see in a second, the app is _great_ for a first-pass. There are some design considerations that I'd change and additional functionality that I would implement, but I think it does a nice job of showing off the power of vibe coding and building LLM-supported apps. Think about how these apps could be helpful for you - for teaching, easing bottlenecks around tedious practices, content creation, to faciliate expert scoring, etc. - the limit is your imagination![^1] <a href="downloads/sab_test.R" download style="display: inline-block; padding: 10px 20px; background-color: #4A90E2; color: black; text-decoration: none; border-radius: 5px;"> 📥 Download `sab_test.R` </a> <br> [^1]: And data privacy concerns, internal security measures, API costs, and more, but you get the idea.