personal-finance

Personal Accounting

Help users record income and expenses, generate spending reports, and manage budgets. All data is stored locally to protect financial privacy.

Use Cases

The user says "Record an entry: lunch 35 yuan" or "Spent 500 on clothes today"
The user says "How much did I spend this month" or "Show me my spending report"
The user says "Set a monthly dining budget of 3000" or "How much budget is left"

Execution Method

Store accounting data in local JSON files, and the LLM parses natural language accounting instructions.

Data Storage

Accounting file:~/Documents/xiaodazi_finance/records.json

{
  "records": [
    {
      "date": "2026-02-09",
      "type": "expense",
      "amount": 35.00,
      "category": "dining",
      "description": "lunch",
      "currency": "CNY"
    }
  ],
  "budgets": {
    "dining": { "monthly_limit": 3000, "currency": "CNY" }
  }
}

Record Entry

# 读取现有记录
cat ~/Documents/xiaodazi_finance/records.json 2>/dev/null || echo '{"records":[],"budgets":{}}'

# 追加记录（通过 Python）
python3 -c "
import json, os
path = os.path.expanduser('~/Documents/xiaodazi_finance/records.json')
os.makedirs(os.path.dirname(path), exist_ok=True)
try:
    data = json.load(open(path))
except:
    data = {'records': [], 'budgets': {}}
data['records'].append({
    'date': '2026-02-09',
    'type': 'expense',
    'amount': 35.00,
    'category': 'dining',
    'description': 'lunch',
    'currency': 'CNY'
})
json.dump(data, open(path, 'w'), ensure_ascii=False, indent=2)
print('recorded')
"

Generate Report

import json
from collections import defaultdict

data = json.load(open("~/Documents/xiaodazi_finance/records.json"))

# 按类别汇总
by_category = defaultdict(float)
for r in data["records"]:
    if r["type"] == "expense":
        by_category[r["category"]] += r["amount"]

for cat, total in sorted(by_category.items(), key=lambda x: -x[1]):
    print(f"{cat}: {total:.2f}")

Expense Categories

Category	Keyword Examples
dining	Lunch, dinner, takeout, coffee
transport	Taxi, subway, fuel, parking
shopping	Clothes, electronics, daily necessities
housing	Rent, utilities, property management
entertainment	Movies, games, travel
education	Courses, books, training
health	Medical care, health checkups, fitness

Security Rules

Local data storage: Financial data is stored only locally and is not uploaded
No automatic sharing: Do not write financial information into other files or send it

Output Specifications

After recording an entry, confirm: category, amount, date
Display reports in tables, sorted by amount in descending order
Proactively remind when the budget is overspent
Support multiple currencies (automatically selected based on user habits)