✍🏻 Installing R Packages from GitHub

August 26, 2019


  • R Studio
  • GitHub and a Personal Access Token


Private Repository prerequisites

Note: If you wish to install from a private repo you need a token from your GitHub Settings. Otherwise feel free to skip this section.

Set a GITHUB_PAT environement variable

Note: Don't commit your tokens!


or in your .Renviron


Note: You will need to sign-in and start a new R session for the environment variable (accessed by Sys.getenv) to be visible.

Required libraries

Baked into the R devtools package there lives a function to install R packages hosted on GitHub.

install.packages("devtools") # Install latest devtools from CRAN
library(devtools)            # Load/Import devtools into working session

Install a package

This exposes the install_github() function for use. Listed below are the arguments we are interested in populating.

repo        # Repository address format username/package
ref         # Commit, tag, or branch name. Default value is Master
auth_token  # User personal access token

Our Final Command should look like this

install_github(repo="username/package", auth_token=Sys.getenv("GITHUB_PAT"))

Note: There are handlers for all flavors of version control.

install_github() from GitHub
install_gitlab() from GitLab
install_bitbucket() from Bitbucket
install_url() from an arbitrary url
install_git() and install_svn() from an arbitrary git or SVN repository
install_local() from a local file on disk
install_version() from a specific version on CRAN

Update a package

update_packages() updates a package to the latest version. This works both on packages installed from CRAN as well as those installed from any of the functions listed above.


Add ref parameter to the function with the specific branch name in which you want to develop. This can be a branch, tag, or commit.

install_github(repo="username/package", ref="develop", auth_token=Sys.getenv("GITHUB_PAT"))

Distributing a package build privately

  • RStudio Shared Directory: Save the R package source code in a /shared-folder/.
  • Users can install these packages by navigation to the toolbar menu and executing.

    • Option 1: Tools > Install Packages > Select Build from picklist
    • Option 2: use the install.packages('<path>') function from the command line.


  • devtools - Tools to make an R developer's life easier
  • github docs - Creating a personal access token for the command line