Code Reviewer Cn

代码审查、重构建议、安全漏洞检查、命名规范、复杂度分析、注释文档生成. Use when you need code reviewer cn capabilities. Triggers on: code reviewer cn, 圈复杂度估算, 嵌套深度检测, 命名风格一致性, 注释率计算与评级, 重复行检测.

Security Scan

The skill is internally consistent for a local code-review helper: it inspects code files, reports metrics, and writes local logs; there are minor mismatches in docs vs scripts and a couple of small safety notes but nothing that indicates misdirection or exfiltration.

Purpose & Capability

Overall the files and scripts implement a code-review tool (complexity, naming, comments, security checks). Minor inconsistencies: SKILL.md lists commands like 'refactor' and 'document' and shows usage 'code-reviewer-cn run', but scripts implement slightly different command names (review.sh supports review/complexity/naming/comments/duplicates/security/all; script.sh exposes init/check/build/test/etc.). This looks like minor documentation/packaging drift rather than malicious behavior.

Instruction Scope

The runtime instructions and the included scripts operate on user-specified files/directories (expected for a code reviewer). They read files, run grep/awk/awk/sort/md5sum/bc and produce stdout reports. The scripts also create and write logs under a per-user data directory. One small concern: check_comments uses eval "$(file_stats "$f")" to import variables; while current file_stats output appears numeric and controlled, using eval with any output derived from file content can be risky if the code is modified—worth reviewing before running on untrusted code or running in a sandbox.

Install Mechanism

No install spec; this is instruction-only with scripts bundled in the skill. No network downloads or external installers are used. Scripts rely on standard UNIX utilities (grep, awk, md5sum, bc, sort, uniq, nl) which is proportionate for the stated functionality.

Credentials

The skill requests no secrets or credentials. It honors an optional CODE_REVIEWER_CN_DIR env var and otherwise uses XDG_DATA_HOME/HOME for a data dir—reasonable for storing usage history/logs. No unrelated environment variables or credentials are required.

Persistence & Privilege

always:false (no forced global presence). The tool persists logs and a data directory under the user's home (~/.local/share/code-reviewer-cn by default). That's normal for a developer tool but users should be aware it writes history.log/history to their home directory.

Assessment

This skill appears to be a local code-review helper and not an exfiltration tool, but take these precautions before installing or running it: 1) Review the bundled scripts (already included) — especially the eval usage in check_comments — or run them in a sandboxed environment. 2) Be aware it will create a data directory (defaults to ~/.local/share/code-reviewer-cn) and append to history.log; set CODE_REVIEWER_CN_DIR to a safe path if desired. 3) It depends on common Unix utilities (md5sum, bc, awk, grep, nl); ensure those are available and from trusted system packages. 4) Avoid running it as root or against highly sensitive directories until you’ve inspected the scripts. Overall the skill is coherent with its stated purpose.

代码审查、重构建议、安全漏洞检查、命名规范、复杂度分析、注释文档生成

核心特点

🎯 精准 — 针对具体场景定制化输出 📋 全面 — 多个命令覆盖完整工作流 🇨🇳 本土化 — 完全适配中文用户习惯

命令速查

  review          review
  refactor        refactor
  security        security
  naming          naming
  complexity      complexity
  document        document

专业建议

• 边界条件处理（空值、零、负数、最大值）
• 错误处理（try-catch、错误码）
• 并发安全（竞态条件、死锁）
• 资源释放（文件句柄、数据库连接）
• 避免N+1查询

---

code-reviewer-cn by BytesAgain

💬 Feedback & Feature Requests: https://bytesagain.com/feedback Powered by BytesAgain | bytesagain.com

• Run code-reviewer-cn help for all commands

Commands

Run code-reviewer-cn help to see all available commands.

When to Use

• Quick code tasks from terminal
• Automation pipelines

Output

Results go to stdout. Save with code-reviewer-cn run > output.txt.