← 返回 Skills 市场

Community Data Process

Name: Community Data Process
Author: yuhui435

作者 yuhui435 · GitHub ↗ · v1.0.0 · MIT-0

cross-platform ✓ 安全检测通过

141

总下载

当前安装

版本数

在 OpenClaw 中安装

/install community-data-process

功能描述

北汽社群数据导出：数据清洗 → 数据校对 → 数据合并。从客户群导出文件中筛选温冷一期和试点店标签数据，校对关键指标一致性，增量合并到 BI_社群数据上传表。统计日期使用源文件创建时间（下载日期）。每天 00:55 自动执行。

使用说明 (SKILL.md)

北汽社群数据导出

每日社群数据导出 → 清洗 → 校对 → 合并到 BI 系统。

流程概览

1. 数据清洗 → 2. 数据校对 → 3. 数据合并 → 4. 最终验证

使用方式

一键执行全流程

python ~/.openclaw/workspace-pm/skills/community-data-process/run.py

分步执行

# 第 1 步：数据清洗
python ~/.openclaw/workspace-pm/skills/community-data-process/run.py clean

# 第 2 步：数据校对
python ~/.openclaw/workspace-pm/skills/community-data-process/run.py audit

# 第 3 步：数据合并
python ~/.openclaw/workspace-pm/skills/community-data-process/run.py merge

# 第 4 步：最终验证
python ~/.openclaw/workspace-pm/skills/community-data-process/run.py verify

规则

清洗规则

源文件：Downloads 目录下按**创建时间（下载时间）**排序，取最新的 客户群导出*.xlsx
筛选条件：O 列（群标签）= 温冷一期或试点店
数字列格式转换：群人数、员工人数、客户人数、今日入群、今日退群、今日消息 → int

校对规则

三方对比：源文件 vs 清理后 vs 合并后
8 个指标全部一致才通过
数据质量：无空值、无负值

合并规则

模式：增量添加（不去重）
列映射：源文件 A-O 列 → 目标文件 E-S 列
D 列统计日期 = 源文件的创建时间（下载日期），文件是哪天下载的就填哪天（不是脚本执行日期）
A-C 列留空

列映射

源文件	→	目标文件	列名
A	→	E	群 ID
B	→	F	群名称
C	→	G	群主
D	→	H	群管理员
E	→	I	群人数
F	→	J	群活跃
G	→	K	群类型
H	→	L	员工人数
I	→	M	客户人数
J	→	N	今日入群
K	→	O	今日退群
L	→	P	今日消息
M	→	Q	入群时间
N	→	R	最后发言时间
O	→	S	群标签

输出文件

文件	说明
`客户群导出_清理后_温冷一期 + 试点店_YYYYMMDD.xlsx`	清洗后数据
`BI_社群数据上传_已更新_YYYYMMDD.xlsx`	合并后最终文件
`数据校对报告_YYYYMMDD.txt`	校对报告

常见问题

Q1: 今日退群数据不一致

检查是否使用了正确的源文件。不同日期导出的文件数据不同。

Q2: 统计日期错误

脚本自动读取源文件的下载日期，不需要手动指定。

Q3: 数字列显示为文本

脚本自动转换 6 个数字列为 int 格式。

依赖

Python 3.11+
pandas
openpyxl

安全使用建议

This skill appears to do what it says: local Excel processing (filter → audit → merge) with no network calls or credential access. Before installing or enabling autonomous runs, test it on copies of your files: it uses the Windows-style Downloads path and selects columns by index (position), so mismatched column orders or non-Windows environments may cause errors or incorrect mappings. It appends rows without de-duplication, so running multiple times can create duplicates — keep backups of the BI file. Also note it writes a small result file (.baic_community_result.txt) to Downloads. If you plan to run it automatically, schedule only after confirming mappings are correct and that the Downloads folder contains the expected filenames and column layout.

功能分析

Type: OpenClaw Skill Name: community-data-process Version: 1.0.0 The skill is a legitimate data processing automation for BAIC community Excel reports, performing data cleaning, auditing, and merging. The code in `run.py` strictly follows the business logic described in `SKILL.md`, operating only on files within the user's Downloads directory without any signs of data exfiltration, malicious execution, or unauthorized persistence.

能力评估

✓ Purpose & Capability

Name/description describe exporting, cleaning, auditing and merging Excel data from customer-group exports; the included run.py implements precisely those actions (find latest export in Downloads, filter by labels, validate numeric columns, map and append rows to BI file). No unrelated credentials, binaries, or services are requested.

✓ Instruction Scope

SKILL.md and run.py both operate on local Excel files in the user's Downloads folder and on a local BI Excel file. The script uses file creation time as the 'download date', reads/writes only local files, and writes a small result text file to Downloads. There is a minor robustness issue (column selection by index and a hard-coded Windows-style Downloads path) that could cause incorrect behavior if source files differ, but it is scope-consistent rather than malicious.

✓ Install Mechanism

No install spec and no external downloads — instruction-only plus a Python script. Dependencies are standard (Python 3.11+, pandas, openpyxl). Nothing is fetched from arbitrary URLs or installed system-wide.

✓ Credentials

The skill requires no environment variables, credentials, or config paths. The script exclusively accesses local files in the Downloads directory and writes output there; requested access is proportional to the stated file-processing task.

✓ Persistence & Privilege

always:false (default) and no mechanism to persist or modify other skills or global agent settings. The script writes its own result file in Downloads but does not alter other skills or request elevated privileges. Autonomous invocation is allowed by platform default but does not combine with other privilege escalations here.

如何使用

确保已安装 OpenClaw（本地或 Docker 部署）
在对话框中输入安装命令：/install community-data-process
安装完成后，直接呼叫该 Skill 的名称或使用 /community-data-process 触发
根据 Skill 的参数说明提供必要输入，即可获得结构化输出

版本历史

v1.0.0

Initial release of community-data-process skill: - Automates daily export, cleaning, validation, and merging of community group data to the BI system. - Implements multi-step process: data cleaning, auditing, merging, and final verification. - Filters specific group labels, checks multiple key indicators for consistency, and handles incremental merging. - Uses file creation date as the statistics date for all records. - Provides detailed usage instructions, rules, column mappings, expected outputs, and FAQ. - Scheduled to run automatically every day at 00:55.

元数据

Slug community-data-process

版本 1.0.0

许可证 MIT-0

累计安装 1

当前安装数 1

历史版本数 1

常见问题