review-r

name	review-r
description	R code review on existing scripts. Checks quality, reproducibility, domain correctness, professional standards. Produces report without editing files. Use when evaluating or auditing existing R code, not writing new analysis.
when_to_use	review R script, check R code, replication-ready, audit R file, follow conventions, will this reproduce, check analysis script, code review, review-r, quality feedback on .R file
argument-hint	[filename, 'all', or analysis name pattern]
allowed-tools	["Read","Grep","Glob","Write","Task","AskUserQuestion"]

Review R Scripts

Run the comprehensive R code review protocol.

Steps

1. Identify scripts to review:

   • If $ARGUMENTS is a specific .R filename: review that file only
   • If $ARGUMENTS is a name pattern (e.g., model_name): glob for matching .R files. If multiple matches, use AskUserQuestion:
     • header: "Scripts"
     • question: "Multiple R scripts match that pattern. Which should I review?"
     • multiSelect: true
     • options: list up to 4 matched files (label: filename, description: path and last modified). User can select multiple.
   • If $ARGUMENTS is all: review all R scripts in scripts/R/ and Figures/*/
   • If $ARGUMENTS is empty, glob for all .R files. If multiple found, use AskUserQuestion as above.

2. For each script, launch the r-reviewer agent with instructions to:

   • Follow the full protocol in the agent instructions
   • Read rules/r-code-conventions.md for current standards
   • Save report to quality_reports/[script_name]_r_review.md

3. After all reviews complete, present a summary:

   • Total issues found per script
   • Breakdown by severity (Critical / High / Medium / Low)
   • Top 3 most critical issues
   • For every Critical or High issue, include a structured category: execution, data_provenance, identification, manuscript_handoff, artifact_size, style, or other
   • State whether each severe issue is transient, whether a caveated draft is allowed, whether new data or new analysis is required, and the recommended disposition: rerun_analysis, rerun_data, proceed_with_caveats, or stop_for_attention

4. IMPORTANT: Do NOT edit any R source files. Only produce reports. Fixes are applied after user review.