Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →brycewang-stanford/r-reproducibility-guide
Create reproducible research workflows with R and RMarkdown/Quarto
Passed 20 of 22 test cases with the skill, and 20 of 22 without it.
1 fixed19 passed either way1 still failing1 broke
Tokens, turns and time describe resource use, not a dollar price.
A test case the skill fixed. Same model, same prompt, with and without the skill.
I am setting up the initial setup chunk of an RMarkdown file for a research article. I want global knitr options that display code chunks, suppress warnings, set figure width to 7 inches, and set image resolution to 300 DPI. What R chunk configuration code should be in opts_chunk$set()?
| Checked | Without skill | With skill |
|---|---|---|
| Configures knitr options with message = FALSE. | Missed | Passed |
| Configures knitr options with warning = FALSE. | Passed | Passed |
| Sets figure resolution using dpi = 300. | Passed | Passed |
The instructions your model reads when the skill loads.
A skill for creating fully reproducible research workflows in R using RMarkdown, Quarto, package management with renv, and project organization best practices. Covers literate programming, environment management, automated reporting, and sharing reproducible analyses.
my-research-project/
README.md
my-project.Rproj # RStudio project file
renv.lock # Package versions (managed by renv)
renv/ # renv library directory
data/
raw/ # Untouched original data
processed/ # Cleaned, analysis-ready data
R/
01-clean.R # Data cleaning functions
02-analyze.R # Analysis functions
03-visualize.R # Plotting functions
utils.R # Helper functions
analysis/
main-analysis.Rmd # Primary analysis notebook
supplementary.Rmd # Supplementary analyses
output/
figures/ # Generated plots
tables/ # Generated tables
manuscript.pdf # Compiled document
Makefile # Reproducible build commands
1. Raw data is read-only (never modify original data files)
2. All processing steps are scripted (no manual spreadsheet edits)
3. Generated outputs can be deleted and recreated from source
4. Package versions are locked with renv
5. Random seeds are set for all stochastic operations
test cases passed, vs 20 / 22 without
More skills from the same source and nearby in the registry, with their headline benchmark lift.