r-skills

Use when designing R function APIs, reviewing R code for design issues, writing functions for R packages, or evaluating argument ordering and naming decisions. Does NOT cover: style/linting, error handling (rlang-conditions), CLI output (r-lib:cli), testing (testing-r-packages), CRAN compliance (cran-extrachecks).

Use when working with R ggplot2 package, especially ggplot2 4.0+ features. Covers S7 migration (@ property access), theme defaults with ink/paper/accent, element_geom(), from_theme(), theme shortcuts (theme_sub_*), palette themes, labels with dictionary/attributes, discrete scale improvements (palette, continuous.limits, minor_breaks, sec.axis), position aesthetics (nudge_x/nudge_y, order), facet_wrap dir/space/layout, boxplot/violin/label styling, stat_manual(), stat_connect(), coord reversal.

Use when creating an R modeling package that needs standardized preprocessing for formula, data frame, matrix, and recipe interfaces. Covers: mold() for training data preprocessing, forge() for prediction data validation, blueprints, model constructors, spruce functions for output formatting.

Use when writing R code that manipulates expressions, builds code programmatically, or needs to understand rlang's defuse/inject mechanics. Covers: defusing with expr()/enquo()/enquos(), quosure environment tracking, injection with !!/!!!/{{, symbol construction with sym()/syms(). Does NOT cover: data-mask programming patterns (tidy-evaluation), error handling (rlang-conditions), function design (designing-tidy-r-functions).

Use when developing R packages that need to handle errors, warnings, and conditions properly. Covers: (1) cli_abort/cli_warn/cli_inform for throwing conditions with formatting, (2) Error call context with caller_env() and caller_arg(), (3) Input validation helpers that report the right function, (4) Error chaining with try_fetch() and parent argument, (5) Testing error conditions with testthat snapshots.

Use when writing complex targets pipelines with branching, batching, or custom target factories. Symptoms: multiple similar targets, parameter sweeps, method comparisons, simulation studies. Covers: static branching (tar_map/tar_combine), dynamic branching (pattern argument), hybrid patterns, custom target factories, pipeline debugging. Does NOT cover: basic targets usage, HPC deployment, package development workflows (see targetopia-packages.md).

Use when programming with tidyverse data-masked functions (dplyr, ggplot2, tidyr) and need to pass column references through functions. Covers: forwarding patterns with {{ and ..., names patterns with .data/.env pronouns, bridge patterns with across()/all_of(), double evaluation and ambiguity pitfalls. Does NOT cover: expression mechanics (r-metaprogramming), error handling (rlang-conditions), function design (designing-tidy-r-functions).

This skill should be used when working with R tidymodels packages, including when the user asks to "create a tidymodels workflow", "build a recipe", "tune a model", "use parsnip", "set up resampling", "create a workflow_set", "compare models", "stack models", or mentions tidymodels packages like recipes, parsnip, workflows, workflowsets, tune, rsample, yardstick, or stacks. Provides ecosystem context before package-specific skills.