The R package softwareRisk leverages the network-like architecture of scientific models together with software quality metrics to identify chains of function calls that are more prone to generating and propagating errors. It operates on tbl_graph objects representing call dependencies between functions (callers and callees) and computes risk scores for individual functions and for paths (sequences of function calls) based on cyclomatic complexity, in-degree and betweenness centrality. The package supports variance-based uncertainty and sensitivity analyses to assess how risk scores change under alternative risk definitions.

Main features

Automatic call-graph construction: call_graph_fun() builds the graph (calls plus cyclomatic complexity) directly from an installed R package or a directory of .R scripts; read_call_graph() imports and validates edge-list and complexity tables prepared with external tools for models in other languages.
Node and path risk scores: all_paths_fun() scores every function and every entry-to-sink call path from cyclomatic complexity, in-degree and betweenness.
Intervention analysis: node_exposure_fun() separates chokepoints from hotspots; path_fix_heatmap() and fix_portfolio_fun() show which fixes, and which sets of fixes under a budget, most reduce path risk.
Uncertainty and sensitivity analysis: uncertainty_fun() propagates alternative risk definitions through the scores (via sensobol); rank_robustness_fun() reports how stable the top-k ranking is; sensitivity_plot_fun() shows which assumption drives the results.
Visualization: plot_top_paths_fun(), path_uncertainty_plot(), rank_robustness_plot() and companions render the results with a publication-oriented theme.

Installation

To install the stable version on CRAN, use

install.packages("softwareRisk")

To install the development version, use devtools:

install.packages("devtools") # if you have not installed devtools package already
devtools::install_github("arnaldpuy/softwareRisk", build_vignettes = TRUE)

Usage

Please see the vignette for a walkthrough of the package utilities.

A printable PDF version of this vignette is also available here.

Citation

To cite softwareRisk in publications:

Puy A (2026). softwareRisk: Computation of node and path-level risk scores in scientific models. R package version 0.3.0, https://github.com/arnaldpuy/softwareRisk.

A BibTeX entry for LaTeX users is

@Manual{puy2026_softwareRisk, title = {softwareRisk: Computation of node and path-level risk scores in scientific models}, author = {Arnald Puy}, year = {2026}, note = {R package version 0.3.0}, url = {https://github.com/arnaldpuy/softwareRisk} }