riskmetric R Package

The {riskassessment} package is a Shiny-based front-end that extends riskmetric. It provides an organizational hub to review and assess R package risks using a local database.

Key features include:

• Storing and displaying all riskmetric metrics and package risk scores over time.
• Managing organization-wide metric weighting and automated decision rules (e.g., whether to endorse or prohibit a package).
• Facilitating communication through package-level user dialogue regarding perceived risk.

The riskmetric workflow for assessing the risk of an R package follows a three-step pipeline:

1. pkg_ref(): Find a source for package information (either an installed package or a CRAN/git source).
2. pkg_assess(): Assess the package against specific validation criteria.
3. pkg_score(): Score the assessment criteria to produce risk metrics.

The final output is a dataset containing validation criteria and an overall risk score for each package.

library(dplyr)
library(riskmetric)

# Example workflow
pkg_ref(c("riskmetric", "utils", "tools")) %>%
  pkg_assess() %>%
  pkg_score()

You can install riskmetric from CRAN or directly from GitHub using devtools.

# From CRAN
install.packages("riskmetric")

# From GitHub
devtools::install_github("pharmaR/riskmetric")

To estimate the risk of one or more packages, use the piped workflow combining pkg_ref(), pkg_assess(), and pkg_score(). This allows you to scrape metadata (locally or remotely), assess it, and derive composite risk measures or individual scores for validation reports.

library(dplyr)
library(riskmetric)

pkg_ref(c("riskmetric", "utils", "tools")) %>%
  pkg_assess() %>%
  pkg_score()