review-r

Reviews R scripts for code quality, reproducibility, and tidyverse conventions without editing source files.

4|Updated Aug 22, 2025
One-click install
npx skills add https://github.com/stevejbickley/arcbita-geek-seminars --skill review-r-stevejbickley
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: review-r
Source: https://github.com/stevejbickley/arcbita-geek-seminars/tree/main/sessions/2026-07-29-ClaudeCodeforAppliedSocialScience/starter/.claude/skills/review-r
Command: npx skills add https://github.com/stevejbickley/arcbita-geek-seminars --skill review-r-stevejbickley

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Researchers writing R analysis code for papers need a consistent, read-only audit of their scripts before submission, but manual review is slow and inconsistent across a project. ## Core Features & Use Cases - Read-only R code review: Audits .R scripts for code quality, reproducibility, domain correctness, tidyverse idioms, and professional standards without modifying any source files. - Flexible scope selection: Review a single file, all scripts for a given lecture, or every script in scripts/R/ and Figures/*/. - Structured severity reporting: Saves a per-script report to quality_reports/ and summarizes issues by severity (Critical / High / Medium / Low) with the top critical findings. - Use Case: Before submitting a paper, run the review across all analysis scripts to catch reproducibility and style issues, then pair with a reproducibility audit for numeric verification. ## Quick Start Ask the assistant to review this R script or run a code review on all R scripts for Lecture 3.

Frequently Asked Questions about review-r

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I review an R script for code quality and reproducibility?▼

Invoke the review with a specific .R filename, a lecture identifier like Lecture3, or 'all' to cover every script in scripts/R/ and Figures/*/. The review checks quality, reproducibility, domain correctness, and tidyverse idioms, then writes a report to quality_reports/.

What does an R code review check before paper submission?▼

The review checks code quality, reproducibility, domain correctness, tidyverse idioms, and professional standards against the project's r-code-conventions rules. Findings are grouped by severity: Critical, High, Medium, and Low.

Does the R code review modify or fix my scripts?▼

No, the review is strictly read-only and never edits R source files. It only produces markdown reports in quality_reports/, and fixes are applied separately after you review the findings.

Can the R review verify that my analysis numbers are correct?▼

No, this review does not run the code or verify numeric outputs. For numeric verification you should pair it with the audit-reproducibility workflow, which executes and checks results.

How do I review all R scripts for a specific lecture at once?▼

Pass the lecture identifier, such as Lecture3, as the argument. The review then audits every R script matching that lecture and produces one report per script plus a combined severity summary.