Can support running shiny app?

[bianchenhao]

claude code cli:

❯ run  #                                                                                                                                                              
  # This is a Shiny web application. You can run the application by clicking                                                                                         
  # the 'Run App' button above.                                                                                                                                          
  #                                                                                                                                                                    
  # Find out more about building applications with Shiny here:                                                                                                           
  #                                                                                                                                                                    
  #    https://shiny.posit.co/                                                                                                                                       
  #                                                                                                                                                                  
                                                                                                                                                                     
  library(shiny)                                                                                                                                                     
                                                                                                                                                                     
  # Define UI for application that draws a histogram                                                                                                                 
  ui <- fluidPage(                                                                                                                                                   
                                                                                                                                                                     
      # Application title                                                                                                                                            
      titlePanel("Old Faithful Geyser Data"),                                                                                                                        
                                                                                                                                                                     
      # Sidebar with a slider input for number of bins                                                                                                               
      sidebarLayout(                                                                                                                                                 
          sidebarPanel(                                                                                                                                              
              sliderInput("bins",                                                                                                                                    
                          "Number of bins:",                                                                                                                         
                          min = 1,                                                                                                                                   
                          max = 50,                                                                                                                                  
                          value = 30)                                                                                                                                
          ),                                                                                                                                                         
                                                                                                                                                                     
          # Show a plot of the generated distribution                                                                                                                
          mainPanel(                                                                                                                                                 
             plotOutput("distPlot")                                                                                                                                  
          )                                                                                                                                                          
      )                                                                                                                                                              
  )                                                                                                                                                                  
                                                                                                                                                                     
  # Define server logic required to draw a histogram                                                                                                                 
  server <- function(input, output) {                                                                                                                                
                                                                                                                                                                     
      output$distPlot <- renderPlot({                                                                                                                                
          # generate bins based on input$bins from ui.R                                                                                                              
          x    <- faithful[, 2]                                                                                                                                      
          bins <- seq(min(x), max(x), length.out = input$bins + 1)                                                                                                   
                                                                                                                                                                     
          # draw the histogram with the specified number of bins                                                                                                     
          hist(x, breaks = bins, col = 'darkgray', border = 'white',                                                                                                 
               xlab = 'Waiting time to next eruption (in mins)',                                                                                                     
               main = 'Histogram of waiting times')                                                                                                                  
      })                                                                                                                                                             
  }                                                                                                                                                                  
                                                                                                                                                                     
  # Run the application                                                                                                                                              
  shinyApp(ui = ui, server = server)                                                                                                                                 
                                                                                                                                                                     

● r-studio - execute_r (MCP)(code: "library(shiny)\n\n# Define UI for application that draws a histogram\nui <- fluidPage(\n    titlePanel(\"Old Faithful Geyser
                            Data\"),\n    sidebarLayout(\n        sidebarPanel(\n            sliderInput(\"bins\",\n                        \"Number of bins:\",\n
                                             min = 1,\n                        max = 50,\n                        value = 30)\n        ),\n        mainPanel(\n          
                             plotOutput(\"distPlot\")\n        )\n    )\n)\n\n# Define server logic required to draw a histogram\nserver <- function(input, output) {\n
                              output$distPlot <- renderPlot({\n        x    <- faithful[, 2]\n        bins <- seq(min(x), max(x), length.out = input$bins + 1)\n         
                            hist(x, breaks = bins, col = 'darkgray', border = 'white',\n             xlab = 'Waiting time to next eruption (in mins)',\n             main
                             = 'Histogram of waiting times')\n    })\n}\n\n# Run the application\nshinyApp(ui = ui, server = server)")                                   
  ⎿  R Error: Can't call `runApp()` from within `runApp()`. If your application code contains `runApp()`, please remove it.     

I Started service in Rstudio-server docker, session info:

> sessionInfo()
R version 4.3.2 (2023-10-31)
Platform: x86_64-pc-linux-gnu (64-bit)
Running under: Ubuntu 22.04.3 LTS

Matrix products: default
BLAS:   /usr/lib/x86_64-linux-gnu/openblas-pthread/libblas.so.3 
LAPACK: /usr/lib/x86_64-linux-gnu/openblas-pthread/libopenblasp-r0.3.20.so;  LAPACK version 3.10.0

locale:
 [1] LC_CTYPE=en_US.UTF-8       LC_NUMERIC=C              
 [3] LC_TIME=en_US.UTF-8        LC_COLLATE=en_US.UTF-8    
 [5] LC_MONETARY=en_US.UTF-8    LC_MESSAGES=en_US.UTF-8   
 [7] LC_PAPER=en_US.UTF-8       LC_NAME=C                 
 [9] LC_ADDRESS=C               LC_TELEPHONE=C            
[11] LC_MEASUREMENT=en_US.UTF-8 LC_IDENTIFICATION=C       

time zone: Etc/UTC
tzcode source: system (glibc)

attached base packages:
[1] stats     graphics  grDevices utils     datasets  methods  
[7] base     

other attached packages:
[1] ClaudeR_0.2.0 shiny_1.11.1 

loaded via a namespace (and not attached):
 [1] vctrs_0.6.5       cli_3.6.5         rlang_1.1.6      
 [4] promises_1.3.3    generics_0.1.4    xtable_1.8-4     
 [7] jsonlite_2.0.0    glue_1.8.0        colorspace_2.1-0 
[10] htmltools_0.5.8.1 httpuv_1.6.13     sass_0.4.9       
[13] scales_1.3.0      grid_4.3.2        jquerylib_0.1.4  
[16] munsell_0.5.0     tibble_3.3.0      fastmap_1.2.0    
[19] base64enc_0.1-3   lifecycle_1.0.4   memoise_2.0.1    
[22] compiler_4.3.2    miniUI_0.1.1.1    dplyr_1.1.4      
[25] Rcpp_1.1.0        pkgconfig_2.0.3   rstudioapi_0.17.1
[28] later_1.4.2       digest_0.6.34     R6_2.6.1         
[31] tidyselect_1.2.1  pillar_1.11.0     magrittr_2.0.3   
[34] bslib_0.8.0       tools_4.3.2       gtable_0.3.4     
[37] mime_0.13         cachem_1.1.0      ggplot2_3.5.2 

[IMNMV]

Thanks for reaching out.

This is a bit tricky given claudeAddin() is a shiny app and you can't nest runApp() inside another. With that being said, there is a workaround the current package can do.

As the package stands right now, you could have the agent define your ui and sever objects and write the app.R file but then NOT call shinyApp(). You would just launch it yourself from the RStudio console or via the Run App button.

But, if that's too much effort I could see about adding support for launching Shiny apps in a background process (it would be a separate R session on a different port), so the agent can start the app and return you the URL directly. If you think this approach provides the most utility I'll add it in.

[bianchenhao]

Thanks for reaching out.

This is a bit tricky given claudeAddin() is a shiny app and you can't nest runApp() inside another. With that being said, there is a workaround the current package can do.

As the package stands right now, you could have the agent define your ui and sever objects and write the app.R file but then NOT call shinyApp(). You would just launch it yourself from the RStudio console or via the Run App button.

But, if that's too much effort I could see about adding support for launching Shiny apps in a background process (it would be a separate R session on a different port), so the agent can start the app and return you the URL directly. If you think this approach provides the most utility I'll add it in.

Thanks, here's the background: I want to use Claude Code with vibe coding to build a Shiny app. Although claude code can use shinytest2 to test whether the functionality works correctly during testing, it can't test whether the front-end layout is rendering properly. So I'm hoping to be able to run the Shiny app and read the HTML/CSS etc. Let claude code determine whether the front-end is displaying correctly.

[IMNMV]

Interesting problem!

I tested a possible solution out with Claude and this ended up working locally:

library(shinytest2)

# Launch app in headless browser
app <- AppDriver$new(app_dir = "path/to/app")

# Inspect the rendered HTML/CSS
app$get_html("body")

# Take a screenshot
app$get_screenshot("screenshot.png")

# Clean up when done
app$stop()

The agent can call all of this through execute_r. get_html() returns the full rendered DOM directly, so the agent can inspect layout, CSS classes, etc.

For screenshots, since you're running RStudio Server in Docker, Claude Code won't have direct filesystem access to read the PNG. You can base64-encode it and return it through execute_r instead:

app$get_screenshot("screenshot.png")
base64enc::base64encode("screenshot.png")

The agent can then decode and analyze the image. If you were running locally, it could just read the file directly. However, be careful with context since encoding a base64 image can output a LOT of tokens.

Let me know if that ends up working for you.