CSV Wizard

交互式数据清洗 CLI，支持自动类型推断、缺失值处理、重复检测

Security Scan

The skill's described behavior (a CLI CSV cleaner) is reasonable and requests no secrets, but the package lacks any executable or install instructions—skill metadata claims a bin/clean-csv entry that is not present—so the bundle is internally inconsistent and should be verified before use.

Purpose & Capability

Name/description match a simple CSV-cleaning CLI and do not ask for credentials or unrelated resources, which is appropriate. However, skill.yaml declares an entry point (bin/clean-csv) and README suggests a GitHub repo, but no binary, install spec, or source is included in the package. That mismatch (claiming a CLI but not providing it) is incoherent.

Instruction Scope

SKILL.md only instructs running the /clean-csv command with CSV-related flags (info, preview, fill-missing, drop-duplicates, etc.). There are no instructions to read unrelated system files, access env vars, or send data to external endpoints.

Install Mechanism

There is no install spec and no code files/binaries included despite the skill manifest pointing to an entry binary. README mentions cloning a GitHub repo, but the skill metadata lists Source: unknown and Homepage: none. Absence of a clear, included install mechanism or a trusted release URL is a packaging/integrity risk.

Credentials

The skill declares no required environment variables, credentials, or config paths—appropriate for a local CSV cleaning utility. No hidden env access is visible in the provided instructions.

Persistence & Privilege

The skill does not request always:true and uses default invocation settings. It does not claim to modify other skills or system configuration. No elevated persistence is requested.

What to consider before installing

This skill's functionality (CSV cleaning) appears reasonable and it does not request secrets, but the package is missing the CLI implementation it advertises. Before installing or running anything: 1) do not run unknown binaries; verify the actual executable/source code—follow the README git URL and inspect the repository and releases on GitHub to ensure bin/clean-csv exists and is trustworthy; 2) prefer packages with explicit install specs or known release URLs (GitHub Releases, official domain); 3) if you must try it, run in a sandboxed environment and review the code first; 4) contact the publisher for a canonical source or updated package if you want to proceed. The current bundle is inconsistent and should be treated with caution.

CSV Wizard — 交互式数据清洗工具

强大的 CSV 数据清洗和转换工具，提供自动类型推断、缺失值处理、重复检测等功能。

功能特性

• 🔍 自动类型推断（数字、日期、布尔值、分类变量）
• 🧹 缺失值检测与多种填充策略
• 📊 数据统计摘要与预览
• 🔄 重复行检测与删除
• 📝 列名标准化与重命名
• 🎯 数据格式转换与导出

使用方法

基本清洗

/clean-csv data.csv --output clean-data.csv

交互式清洗（推荐）

/clean-csv data.csv --interactive

预览数据信息

/clean-csv data.csv --info

处理缺失值

/clean-csv data.csv --fill-missing mean --output result.csv

删除重复行

/clean-csv data.csv --drop-duplicates --output result.csv

选项说明

选项	说明
--info	显示数据基本信息
--preview	预览前 N 行数据
--fill-missing	缺失值填充策略（drop/mean/median/mode/constant）
--drop-duplicates	删除重复行
--standardize-names	标准化列名（snake_case）
--interactive	交互式模式
--output	输出文件路径

缺失值填充策略

• drop - 删除包含缺失值的行
• mean - 使用列均值填充（仅数值列）
• median - 使用中位数填充（仅数值列）
• mode - 使用众数填充
• constant - 使用固定值填充（需配合 --fill-value）

示例

# 查看数据摘要
/clean-csv sales.csv --info

# 清洗数据：删除重复行 + 填充缺失值
/clean-csv sales.csv --drop-duplicates --fill-missing median --output clean-sales.csv

# 交互式清洗
/clean-csv data.csv --interactive

# 仅标准化列名
/clean-csv data.csv --standardize-names --output output.csv

输出说明

清洗后的 CSV 文件将保留原始格式，同时：

• 自动处理编码问题（统一为 UTF-8）
• 移除首尾空格
• 标准化布尔值（True/False）