📦 数据

v1.0.3

利用 AI 字段检测清洗并去重多格式数据，实现格式标准化、多源合并，输出 Excel、CSV 或 Feishu Bitable。

0· 31·0 当前·0 累计

by @billjamno58 (YK-Global)·MIT

文件处理数据分析数据库代码审查即时通讯

下载技能包

License

MIT

最后更新

2026/4/24

安全扫描

VirusTotal

无害

查看报告

OpenClaw

可疑

medium confidence

该包声明的元数据与运行时指令在所需凭据及 Feishu 集成上不一致；该 skill 会调用外部服务（计费 + AI），而 SKILL.md 对发送的数据集内容描述模糊——安装或提供密钥前请审查。

评估建议

This 技能 includes 运行nable Python code that will call external 服务s: a billing 端点 (技能pay.me) and AI APIs (OPENAI_API_BASE) for field 检测ion/classification. Before 安装ing or providing 凭证s: - 验证 the publisher and source: registry metadata shows "Source: unknown" and no homepage. Prefer packages with a known mAIntAIner and 仓库. - Ask the author to explAIn the 凭证 gap: the registry shows no required env vars but 技能.md requires OPENAI_API_KEY, 技能_BILLING_API_KEY and FEISHU_USER_ID. Also ask how Feishu writ...

详细分析 ▾

⚠ 用途与能力

The 技能 advertises Feishu Bitable write-back and AI-driven features, which plausibly require external 凭证s, but the registry metadata at the top 列出s "Required env vars: none" and "Primary 凭证: none" — inconsistent with 技能.md and the included code. The 技能.md 列出s OPENAI_API_KEY, 技能_BILLING_API_KEY and FEISHU_USER_ID; the code implements billing (技能pay.me) and AI field identification/classification, so those 环境 variables make sense — but Feishu write-back would normally require Feishu API 凭证s (应用令牌/命令行工具ent secret) which are not declared. This mismatch between clAImed metadata and actual requirements is a red flag.

⚠ 指令范围

运行time invokes scripts/mAIn.py which orchestrates parsing, 清理ing, AI classification, billing, 报告ing and Feishu 输出. The 技能.md and code indicate that data will be sent to AI APIs (OPENAI_API_BASE / OPENAI_API_KEY) for field identification/classification but the doc does not explicitly describe what parts of the user's data设置 are transmitted or how sensitive fields are handled. Billing transmits the FEISHU_USER_ID to 技能pay.me. The instructions grant broad discretion (AI auto-检测ion, semantic inference) that could cause sensitive PII to be sent to external AI 端点s; this is not clearly documented or consented in the 技能.md.

✓ 安装机制

There is no external 安装 spec or 下载s; the 技能运行s local Python scripts bundled in the package. No evidence of fetching code from untrusted URLs or executing remote 安装ers. That reduces supply-chAIn risk compared to arbitrary 下载s, but the included code will 执行 locally and may make outbound network calls.

⚠ 凭证需求

技能.md 请求s multiple sensitive 环境 variables: OPENAI_API_KEY / OPENAI_API_BASE (AI inference), 技能_BILLING_API_KEY (技能Pay billing), 技能_BILLING_技能_ID and FEISHU_USER_ID. These map to the 技能's features, but they are not 列出ed in the registry metadata (incoherent). 添加itionally, Feishu write-back capability would normally require Feishu API 凭证s (应用令牌s / 访问令牌s), but no such variables are declared; either the feature is incomplete or it expects implicit 平台-level Feishu 凭证s (not disclosed). Billing.py also implements a "dev mode" where missing billing API key or network errors cause the code to 'fAIl open' and allow operation without charge; that behavior should be explicit to users because it affects billing correctness and abuse risk.

✓ 持久化与权限

The 技能 does not 请求 always:true and does not clAIm to modify other 技能s or global 代理设置tings. It 运行s as a subprocess (scripts/mAIn.py). Autonomous invocation is allowed (平台 default) but not combined with excessive declared privileges here; that's acceptable but be aware an autonomously-invoked 技能 can still exfiltrate data if given API keys.

安全有层次，运行前请审查代码。

License

MIT

查看条款 ↗

运行时依赖

无特殊依赖

版本

latestv1.0.32026/4/24

- 本次版本未检测到代码或文档变更。 - 版本号升至 1.0.3，无任何文件或功能改动。 - 功能、用法及计费均无更新。 - 现有功能与文档保持不变。

● 无害

安装命令

点击复制

官方npx clawhub@latest install data-cleaner-ai

镜像加速npx clawhub@latest install data-cleaner-ai --registry https://cn.longxiaskill.com

技能文档

上传凌乱数据 → 获得干净结构化结果。支持多格式解析、AI字段识别、智能去重/补全/格式化、多源合并、飞书原生输出（Bitable + 质量报告文档）。

典型场景： 电商订单清理、CRM客户数据清洗、银行对账、花名册整理、多系统数据合并。

---

能力

F1 · 多格式解析

Excel (.xlsx / .xls)
CSV / TSV
JSON（半结构化）
剪贴板文本粘贴

F2 · 智能字段识别

AI自动识别：姓名、手机、邮箱、地址、金额、日期、SKU、订单号、身份证号、性别等
支持用户自定义字段映射覆盖

F3 · 数据清洗

去重：精确匹配 + 模糊去重（FuzzyWuzzy，阈值88%）
缺失值填充：均值/众数/语义推理/留空
格式标准化：

- 手机 → 1xx-xxxx-xxxx - 日期 → YYYY-MM-DD - 金额 → 保留2位小数 - 地址 → 省/市/区/街道标准化

F4 · 数据分类/打标签（PRO）

8条内置业务规则（高价值客户、沉睡用户、VIP、企业客户等）
支持自定义JSON规则
AI自动打标签（需PRO + AI API Key）

F5 · 多源合并（PRO）

按关键字段跨文件关联合并
无精确键时启用模糊合并（FuzzyWuzzy）
冲突字段解决：按源顺序或最新时间戳优先

F6 · 飞书原生输出

Excel / CSV 导出
飞书Bitable（多维表）回写
数据质量报告自动生成飞书文档（Markdown）

---

功能矩阵

| 功能 | FREE | PRO | |------|:----:|:---:| | 多格式解析 | ✅ | ✅ | | 基础去重 | ✅ | ✅ | | 智能填充 | ❌ | ✅ | | 格式标准化 | ❌ | ✅ | | 模糊去重 | ❌ | ✅ | | 多源合并 | ❌ | ✅ | | AI分类 | ❌ | ✅ | | 数据质量报告 | ❌ | ✅ | | 飞书Bitable输出 | ❌ | ✅ |

---

计费

按次计费（无月费）：

| 版本 | 单次价格 | |------|---------| | FREE | $0.00 USDT | | PRO | $0.01 USDT |

每次清洗或合并流程 = 1次计费。

---

使用

飞书触发

  
data cleaning  
deduplication  
spreadsheet cleanup  
CRM data cleanup  
Excel cleaning

CLI

bash  
python scripts/main.py clean -i data.xlsx -o cleaned.xlsx  
python scripts/main.py clean -t "name,phone\nJohn,13800138000" -f csv -o cleaned.csv  
python scripts/main.py merge --sources customers.xlsx orders.csv --on phone -o merged.xlsx

  
Python API

python  
from main import run_clean_pipeline  
result = run_clean_pipeline(  
    sources=["orders.xlsx"],  
    output_format="xlsx",  
    output_path="/tmp/cleaned.xlsx",  
    dedup_strategy="auto",  
    fill_strategy="auto",  
    classify=True,  
    ai_model="deepseek",  
    generate_report=True,  
)

  
---  
目录结构

`data-cleaner-ai/ ├── SKILL.md └── scripts/ ├── main.py # 入口：run_clean_pipeline / run_merge_pipeline ├── parser.py # F1：多格式解析 ├── field_identifier.py # F2：AI字段识别 ├── cleaner.py # F3：清洗引擎 ├── classifier.py # F4：分类/打标签 ├── merger.py # F5：多源合并 ├── reporter.py # F6：质量报告生成 ├── output.py # F6：输出（Excel/CSV/Bitable/飞书文档） ├── tier_limits.py # 版本权限控制 └── billing.py # SkillPay计费集成`

---

`计费`

本技能使用 SkillPay (skillpay.me) 按次计费。费用：每次执行 $0.0100 USDT（所有付费档）外部API：https://skillpay.me/api/v1/billing传输数据：用户标识（FEISHU_USER_ID环境变量）计费在每次清洗或合并开始时触发；余额不足时返回payment_url 供充值。

---

`必需环境变量`

| 变量 | 说明 | |------|------| |FEISHU_USER_ID| 飞书用户 open_id，用于计费识别 | |OPENAI_API_KEY| AI模型API Key（OpenAI、MiniMax或兼容端点） | |OPENAI_API_BASE| AI API基地址（可选，默认MiniMax端点） | |SKILL_BILLING_API_KEY| skillpay.me 的 Builder API Key（付费调用必需） | |SKILL_BILLING_SKILL_ID | SkillPay上的技能slug（默认 data-cleaner-ai） |

---

`错误处理`

| 错误 | 处理方式 | |------|----------| | 余额不足 | 返回payment_url` 供充值 | | 网络错误 | … |

数据来源：ClawHub ↗ · 中文优化：龙虾技能库