Smart Expense Tracker

智能收支追踪器 — 通过自然语言快速记账，自动分类，生成日/周/月报告和趋势分析。 支持中文自然语言输入("今天午饭花了35元")，数据本地JSON存储保护隐私。 适用于个人财务管理、预算追踪和消费分析场景。

Security Scan

The skill's code, instructions, and storage needs are consistent with a local, privacy-focused expense tracker — it reads/writes a local JSON file and does not request credentials or network access.

Purpose & Capability

The name/description (Chinese expense tracker) match the included scripts and assets: CLI examples in SKILL.md correspond to scripts/expense-tracker.py and scripts/report-generator.py, and data is stored locally under ~/.openclaw/workspace/data/expenses/expenses.json as claimed.

Instruction Scope

Runtime instructions are limited to running the provided Python scripts, editing the local JSON or categories.json for customization, and generating reports. The SKILL.md does not instruct the agent to read unrelated system files, access network endpoints, or exfiltrate data; the Python code uses only the standard library and manipulates the stated local data file.

Install Mechanism

There is no install spec (instruction-only with bundled scripts). No remote downloads or package installs are requested; all code ships with the skill, minimizing install-time risk.

Credentials

The skill requests no environment variables, no credentials, and no config paths beyond the declared local data path. The local file permission hardening (chmod to user-only where possible) is consistent with the privacy claim.

Persistence & Privilege

always is false and the skill does not attempt to modify other skills or global agent configuration. It only creates/uses its own data directory under the user's home; this level of persistence is appropriate for a local tracker.

Assessment

This skill appears coherent and local-only: it stores data under ~/.openclaw/workspace/data/expenses/expenses.json and uses only Python standard library. Before running, you may (1) inspect the two Python scripts to satisfy yourself nothing unexpected is added (they are included), (2) back up any existing expenses.json if you already use that path, and (3) run the scripts in a controlled environment (e.g., a dedicated user account or virtualenv) if you want extra isolation. If you later add integrations that contact external services, re-check required credentials and network calls.

Smart Expense Tracker - 智能收支追踪器

通过自然语言快速记账，自动生成财务报告的智能工具。

功能特性

• 📝 自然语言记账: "今天午饭花了35元"、"收到工资8000"
• 🏷️ 智能自动分类: 自动识别餐饮、交通、购物等分类
• 📊 多维度报告: 日报、周报、月报、趋势分析
• 🔒 本地数据存储: JSON文件存储，保护隐私
• 🌐 中英文支持: 支持中英文自然语言输入
• ⚠️ 消费异常检测: 自动识别异常消费模式

快速开始

添加收支记录

# 记录支出
python scripts/expense-tracker.py add "今天午饭花了35元"
python scripts/expense-tracker.py add "昨天打车去公司28块"
python scripts/expense-tracker.py add "买衣服花了299元"

# 记录收入
python scripts/expense-tracker.py add "今天发工资8000元"
python scripts/expense-tracker.py add "收到红包200"

查看记录

# 列出最近7天记录
python scripts/expense-tracker.py list

# 列出最近30天记录
python scripts/expense-tracker.py list 30

# 只显示支出
python scripts/expense-tracker.py list 7 expense

生成报告

# 日报
python scripts/report-generator.py daily
python scripts/report-generator.py daily 2024-01-15

# 周报
python scripts/report-generator.py weekly
python scripts/report-generator.py weekly 1  # 上周

# 月报
python scripts/report-generator.py monthly
python scripts/report-generator.py monthly 1  # 上月

# 趋势分析
python scripts/report-generator.py trend 30

查看统计

# 最近30天收支摘要
python scripts/expense-tracker.py summary

# 最近90天摘要
python scripts/expense-tracker.py summary 90

# 查看分类列表
python scripts/expense-tracker.py categories

项目结构

smart-expense-tracker/
├── SKILL.md                      # 本文件
├── scripts/
│   ├── expense-tracker.py       # 核心记账逻辑
│   └── report-generator.py      # 报告生成器
└── assets/
    └── categories.json          # 分类配置

数据存储

• 路径: ~/.openclaw/workspace/data/expenses/expenses.json
• 格式: JSON
• 隐私: 纯本地存储，不上传云端

数据结构:

{
  "expenses": [
    {
      "id": "a1b2c3d4",
      "date": "2024-01-15",
      "type": "expense",
      "amount": 35.00,
      "category": "餐饮",
      "note": "午饭",
      "raw_text": "今天午饭花了35元",
      "created_at": "2024-01-15T12:30:00"
    }
  ],
  "categories": {
    "income": ["工资", "奖金", "投资", ...],
    "expense": ["餐饮", "交通", "购物", ...]
  }
}

自然语言支持示例

时间表达

• "今天..." - 今天
• "昨天..." - 昨天
• "前天..." - 前天
• "2024-01-15..." - 指定日期

金额表达

• "花了35元"
• "花了35块"
• "35元"
• "收入5000"

自动分类关键词

分类	触发关键词
餐饮	吃饭、午饭、晚饭、外卖、奶茶、咖啡、火锅
交通	地铁、公交、打车、滴滴、加油、停车、高铁
购物	购物、买衣服、鞋子、包包、化妆品、淘宝、京东
娱乐	电影、游戏、充值、会员、旅游、KTV
居住	房租、房贷、水电、物业费、装修
工资	工资、薪水、发工资
奖金	奖金、年终奖、分红

分类配置

分类配置存储在 assets/categories.json，包含:

• 收入分类: 工资、奖金、投资、兼职、红包、退款、其他收入
• 支出分类: 餐饮、交通、购物、娱乐、居住、医疗、教育、通讯、人情、其他支出
• 预算预警: 每个分类可设置预算阈值
• 关键词映射: 自定义分类识别关键词

自定义分类

编辑 assets/categories.json 添加自定义分类和关键词。

报告类型

日报

• 当日收支总览
• 分类明细
• 交易记录列表

周报

• 本周收支总览
• 日均支出
• 每日明细
• TOP支出分类

月报

• 月度收支总览
• 预算执行情况
• 周度趋势
• 分类统计
• 日均支出

趋势分析

• 消费趋势（上升/下降/稳定）
• 异常消费检测（超过均值2倍）
• 环比变化分析
• 消费建议

高级用法

删除记录

# 删除指定ID的记录
python scripts/expense-tracker.py delete a1b2c3d4

JSON格式输出

# 报告输出JSON格式（便于程序处理）
python scripts/report-generator.py monthly 0 json

批量导入

可通过直接编辑 expenses.json 文件实现批量导入。

使用建议

1. 每日记账: 养成每天记录的习惯，可以在每天晚上花2分钟回顾当天支出
2. 及时分类: 如果自动分类不准确，可以在备注中添加更具体的关键词
3. 定期复盘: 每周查看周报，每月查看月报，了解自己的消费模式
4. 设置预算: 在 categories.json 中为各分类设置预算预警值

故障排除

无法识别金额

• 确保金额格式正确，如 "35元"、"35块"
• 尝试明确表达："花了35元"

分类不准确

• 在描述中添加更多关键词，如 "淘宝买衣服" 会比 "买衣服" 更准确
• 可在 categories.json 中添加自定义关键词

数据文件损坏

• 数据文件是纯JSON，可用文本编辑器修复
• 建议定期备份 ~/.openclaw/workspace/data/expenses/expenses.json

扩展开发

添加新的自动分类规则

编辑 scripts/expense-tracker.py 中的 KEYWORD_MAPPING 字典:

KEYWORD_MAPPING = {
    # 现有规则...
    "新关键词": "目标分类",
    " another keyword": "Category Name"
}

集成其他工具

expense-tracker.py 可作为模块导入:

from scripts.expense_tracker import ExpenseTracker

tracker = ExpenseTracker()
record = tracker.add("今天买书花了50元")
print(record)

技术规格

• Python版本: 3.9+
• 依赖: 无外部依赖，仅使用标准库
• 编码: UTF-8
• 数据格式: JSON

许可证

MIT License