home / skills / deletexiumu / agentskills-hub / wechat-daily-report

wechat-daily-report skill

safe

This skill generates a structured today WeChat report from local data, summarizing stats, to-dos, and amusing anecdotes for clear daily insights.

npx playbooks add skill deletexiumu/agentskills-hub --skill wechat-daily-report

Review the files below or copy the command above to add this skill to your agents.

Files (3)

SKILL.md

3.4 KB

---
name: wechat-daily-report
description: 生成微信聊天今日报告，包含数据统计、待办事项提炼、趣事吃瓜整理。适用于"微信日报"、"今日微信报告"、"聊天记录分析"等场景。
---

# 目标

从本地 DuckDB 读取微信聊天记录，生成结构化的今日报告 Markdown 文件，包含：
- 数据概览（消息总数、活跃聊天、时间分布）
- **待办事项**：按优先级分类，概括提炼需要跟进的工作
- **趣事吃瓜**：整理成完整故事，而非零碎消息
- 幽默总结收尾

# 流程

## 第一步：提取原始数据

运行脚本获取当日聊天数据：

```bash
python3 scripts/extract_data.py [--date YYYY-MM-DD] [--db PATH]
```

脚本会输出：
1. 数据统计（消息数、活跃度、时间分布）
2. 工作群完整对话（用于分析待办）
3. 非工作群对话（用于提取趣事）

## 第二步：概括提炼（AI 完成）

根据原始数据，分析并概括：

### 待办事项提炼规则

1. **识别待办来源**：
   - 被 @淇奥 的消息
   - 包含问题/故障/请求的对话
   - 需要确认/跟进/反馈的事项

2. **按优先级分类**：
   - 🔴 需立即处理：今天要做的、已答应的
   - 🟡 本周跟进：有明确时间节点的
   - 🟢 已解决/知悉：已完成或仅需知悉的

3. **格式要求**：
   ```markdown
   1. **事项标题**
      - 来源：群名/联系人 + 时间
      - 具体内容描述
      - 状态：当前进展
   ```

### 趣事吃瓜提炼规则

1. **识别有趣内容**：
   - 群友热议的话题（多人参与讨论）
   - 包含吃瓜/八卦/搞笑关键词
   - 有完整故事线的对话

2. **整理成故事**：
   - 给每个故事起标题 + emoji
   - 概括事件经过，保留精彩原话
   - 突出笑点/槽点

3. **格式要求**：
   ```markdown
   ### 🍉 故事标题

   故事概括描述，**加粗关键信息**，保留群友精彩原话用引用格式。
   ```

### 今日总结

用一句幽默的话收尾，结合当天的待办和趣事，例如：
> 33 个工作群在线 battle，Git 文件夹还没建，但最重要的是——玩剑网三的红包领不了。

## 第三步：生成报告

将分析结果整理成完整的 Markdown 报告，保存到指定目录。

# 报告模板

```markdown
# 微信日报 (YYYY-MM-DD 周X)

## 一、数据概览
| 指标 | 数值 |
|------|------|
| **消息总数** | X 条 |
| **活跃聊天** | X 个 |
| **活跃人数** | X 人 |
| **峰值时段** | XX:00 (X条) |

## 二、聊天活跃度 TOP 10
（表格）

## 三、消息类型分布
（列表）

## 四、待办事项
### 🔴 需立即处理
### 🟡 本周跟进
### 🟢 已解决/知悉

## 五、趣事吃瓜
### 🍉 故事1
### 📱 故事2
...

## 六、活跃发送者 TOP 10
（表格）

## 七、消息时间分布
（柱状图）

## 今日总结
> 幽默收尾语

---
*报告生成时间: YYYY-MM-DD HH:MM:SS*
```

# 护栏

- 只读操作，不修改数据库
- 待办事项只提取工作相关聊天
- 趣事吃瓜不包含敏感/隐私内容
- 报告默认输出到当前目录

# 资源

- `scripts/extract_data.py`：数据提取脚本
- `scripts/generate_report.py`：基础报告生成脚本（统计部分）

# 使用示例

```bash
# 1. 提取今日数据
python3 scripts/extract_data.py

# 2. AI 根据输出概括提炼，生成最终报告

# 或者直接让 AI 生成（推荐）
# "帮我生成今日微信报告"
```

Overview

This skill generates a structured WeChat daily report in Markdown from local chat data. It summarizes message statistics, extracts and prioritizes work-related todos, organizes entertaining conversation stories, and ends with a short humorous wrap-up. The report is ready to save to the current directory for review or archival.

How this skill works

The skill reads today’s chat records from a local DuckDB snapshot produced by the extraction script, then computes activity metrics and conversation-level groupings. It uses rules to identify work-related follow-ups (mentions, requests, confirmations) and to detect multi-message storylines for fun items. Finally it composes a Markdown report with tables, prioritized todos, curated stories, and a one-line humorous summary.

When to use it

Generate a daily summary of WeChat activity for personal review
Extract and prioritize actionable items from work group chats
Collect and present amusing group conversations as readable stories
Archive daily chat metrics for trends and reporting
Quickly produce a shareable Markdown report of today’s chats

Best practices

Run the data extraction step first to ensure accurate, up-to-date DuckDB input
Limit todo extraction to work groups to avoid false positives from casual chats
Review auto-extracted todos before acting—AI highlights likely follow-ups but may miss context
Filter out or redact any sensitive content when saving or sharing the report
Use the generated Markdown as a baseline; edit titles or priorities to reflect real-world urgency

Example use cases

Morning briefing: generate today’s WeChat report to plan the day’s tasks and meetings
End-of-day recap: capture completed items and unresolved todos for handover
Team lead summary: share activity highlights and pending items from work groups
Social digest: collect top three amusing group stories to post in a newsletter or personal log
Trend tracking: save daily reports to monitor activity spikes and recurring todo patterns

FAQ

Which data source does the skill read from?

It reads chat records from a local DuckDB exported by the data extraction script; the report is read-only and does not modify the database.

How are todos prioritized?

Todos are classified by rules: @mentions and direct requests are high priority, items with explicit dates are medium, and completed or informational items are low priority; the output groups them as 🔴/🟡/🟢.

Can the fun stories include private content?

No. The skill excludes sensitive or private content and only includes stories that are safe to present, summarizing and quoting non-sensitive highlights where appropriate.