跳到主要内容
知仓学习社ZHICANG

review-r

Run the R code review protocol on R scripts. Checks code quality, reproducibility, domain correctness, and professional standards. Produces a report…

不碰外部(只输出文字)无严重或高危命中brycewang-stanford/Auto-Empirical-Research-Skills

它会碰到什么

扫了多少1 个文本文件,1 KB
它会碰到什么不碰外部(只输出文字)
命中总数0 处
命中统计严重 0 · 高 0 · 中 0 · 低 0

这一栏是扫描器报的事实,不是结论。命中多不等于有毒(安全工具、规则库、示例脚本本来就会包含危险写法),命中少也不等于干净。它和你手上的凭据、文件、网络有什么关系,需要你自己看。

技能内容

Review R Code

Run a comprehensive R code review on the specified script(s). Do NOT edit any R source files -- produce a report only.

Steps

  1. Identify target: Use $ARGUMENTS to find the R file(s). If all, scan all .R files in scripts/ and project subdirectories.
  1. Launch r-reviewer agent following the full protocol from .claude/agents/r-reviewer.md.
  1. Read standards from .claude/rules/r-code-conventions.md.
  1. Check 10 categories: Structure & Header, Console Output Hygiene, Reproducibility, Function Design, Domain Correctness, Figure Quality, RDS Data Pattern, Comment Quality, Error Handling, Professional Polish.
  1. Save report to quality_reports/[script_name]_r_review.md.
  1. Present summary: Total issues per script, severity breakdown (Critical/High/Medium/Low), top 3 critical issues.

Important

  • NEVER edit source files. Report only.
  • Prioritize correctness over style.

想直接用这个技能?

本站把开放许可(MIT / Apache 等)的技能按仓库打包整理到网盘,点一下转存到你自己的网盘,不用一个个从 GitHub 拉。许可未声明的技能只给原始仓库链接,不打包。

同名技能的其他版本

有 5 个不同仓库或目录里都有叫 review-r 的技能。它们内容并不相同,别混用: