CRAN_Status_Badge CRAN_Downloads Travis-CI Build Status

Easily add help documentation to shiny elements, using markdown files.

The advantages of using this package are:


shinyhelper 0.3.1 now on CRAN!

You can install the package with:


To get the latest development version, you can use the devtools package to install from GitHub directly:


In both cases, then load the package in with:



There is a live demo hosted on Click here to go to the demo!

Alternatively, run the demo locally with:




You can add help files to any shiny element, including all inputs and outputs, with a simple call to helper():

# load the package

# For elements in your ui you wish to add a help icon to
helper(plotOutput(outputId = "plot"))

# if you have %>% loaded, you can do plotOutput(outputId = "plot") %>% helper()

# In your server script, include:

# this triggers the modal dialogs when the user clicks an icon
# specify the name of your directory of help markdown files here

# e.g. observe_helpers(help_dir = "help_mds") will look for a directory called help_mds

# If you wish to include mathematical formulae in your markdown, use the `withMathJax` argument:
# observe_helpers(withMathJax = TRUE)

You can define helpers in dynamic UI elements as well, and have the help file rendered dynamically on the server side. This allows for :


All you need now is some content for your help page. You can specify this in 2 ways:


To specify inline content, simply set type = "inline" in helper, and supply the title and content arguments. content can be a character vector, in which case each element will be a new line. You can also use raw HTML tags to format your inline content E.g.

plotOutput(outputId = "plot") %>% helper(type = "inline",
                                         title = "Plot",
                                         content = c("This is a <b>plot</b>.",
                                                     "This is on a new line."))


To use markdown, set type = "markdown" in helper, and supply the name of your markdown file (without the .md) in the content argument. This file should be in the directory specified by the help_dir argument to observe_helpers. E.g.

plotOutput(outputId = "plot") %>% helper(type = "markdown",
                                         content = "Plot")

# this will search for '' in the directory given in observe_helpers

You can specify a title argument too, or leave it blank and use a ## Heading in your markdown document.

Changing the Icon Appearance

You can change the type of icon used and its colour, as well as passing CSS inline.


The icons are shiny::icon("question-circle") icons by default, but you can change them individually using the icon argument of helper():

plotOutput(outputId = "plot") %>% helper(icon = "exclamation")

Please see Font Awesome for the available icons.


You can change the icon colour with the colour argument. Pass it any valid CSS colour as a character string.

plotOutput(outputId = "plot") %>% helper(colour = "green")

Inline CSS

You can pass a style argument to modify CSS inline. This applies to the <div> containing the icon.

plotOutput(outputId = "plot") %>% helper(style = "color: red;")

Note: Passing a colour in a style argument will override colour.

Changing the Help Page Size

By default, all help files are medium sized modalDialog() boxes (size = "m"). You can change each one though, by passing the size argument to helper():

plotOutput(outputId = "plot") %>% helper(size = "l")

Other Arguments

You can also change:

Creating your Help Files

There is also a function, create_help_files() to quickly create a directory of help files from a vector of names.

# Run this interactively, not in a shiny app
create_help_files(files = c("Clusters", "Columns", "PlotHelp"), 
                  help_dir = "helpfiles")

The help_dir will be “helpfiles” by default.


Obviously, this package would not be possible (or indeed meaningful) without the incredible shiny package. Full credit to the authors of shiny for doing all of the actual work!

Many thanks also to Guangchang Yu for the wonderful hexSticker package!