openclaw 网盘下载
OpenClaw

技能详情(站内镜像,无评论)

首页 > 技能库 > 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

数据与表格

作者:antonia huang @antonia-sz

许可证:MIT-0

MIT-0 ·免费使用、修改和重新分发。无需归因。

版本:v1.0.1

统计:⭐ 0 · 146 · 0 current installs · 0 all-time installs

0

安装量(当前) 0

🛡 VirusTotal :良性 · OpenClaw :可疑

Package:antonia-sz/csv-wizard

安全扫描(ClawHub)

  • VirusTotal :良性
  • OpenClaw :可疑

OpenClaw 评估

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.

目的

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.

说明范围

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.

安装机制

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.

证书

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.

持久

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.

安装(复制给龙虾 AI)

将下方整段复制到龙虾中文库对话中,由龙虾按 SKILL.md 完成安装。

请把本段交给龙虾中文库(龙虾 AI)执行:为本机安装 OpenClaw 技能「CSV Wizard」。简介: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。
请 fetch 以下地址读取 SKILL.md 并按文档完成安装:https://raw.githubusercontent.com/openclaw/skills/refs/heads/main/skills/antonia-sz/csv-wizard/SKILL.md
(来源:yingzhi8.cn 技能库)

SKILL.md

打开原始 SKILL.md(GitHub raw)

---
name: csv-wizard
description: 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
author: openclaw
version: 1.0.0
commands:
  /clean-csv: 清洗和转换 CSV 文件
---

# 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

```bash
/clean-csv data.csv --output clean-data.csv
```

### 交互式清洗(推荐)

```bash
/clean-csv data.csv --interactive
```

### 预览数据信息

```bash
/clean-csv data.csv --info
```

### 处理缺失值

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

### 删除重复行

```bash
/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

```bash
# 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)