mcpField
M
monthly-digest

monthly-digest

github.com/openaccountantData & databases103424 installsUpdated 29w ago
Repository

About monthly-digest

monthly-digest is a financial-analysis skill for people who want a concise monthly view of income, spending, savings, and unusual activity. It compares the month with the previous one and can export the digest as Markdown.

Key features

  1. 01Monthly income and expense totals
  2. 02Savings rate calculation
  3. 03Category and month-over-month trends
  4. 04Largest transactions and income entries
  5. 05Unusual spending and recurring-charge flags
  6. 06Optional Markdown export

Use cases

  1. 01Review the previous month’s financial position
  2. 02Compare spending categories with the prior month
  3. 03Identify unusually large or recurring charges

MCP support

The skill uses Wilson tools `spending_summary`, `transaction_search`, and `anomaly_detect` to gather and analyze financial data.

Install

Copy the command and run it in a terminal.

$npx skills add https://github.com/openaccountant/skills --skill monthly-digest

After installing, type/monthly-digestin Claude Code to confirm it loaded.

Details

From the project README

Monthly Digest

Overview

Produce a concise monthly financial digest covering income, expenses, savings rate, notable transactions, spending anomalies, and month-over-month trends. Designed to be a quick "state of your finances" snapshot.

Wilson Tools Used

  • spending_summary — category-level spending totals for the month
  • transaction_search — find largest transactions and income entries
  • anomaly_detect — flag unusual spending patterns or amounts

Workflow

  1. Determine the target month (default: previous calendar month).
  2. Use spending_summary to get category totals for the month.
  3. Use spending_summary for the prior month to calculate month-over-month changes.
  4. Use transaction_search to find:
    • Top 5 largest expenses
    • All income transactions
    • Any new vendors (first-time transactions)
  5. Use anomaly_detect to flag:
    • Unusual amounts (significantly above category average)
    • New recurring charges
    • Missing expected recurring charges
  6. Compile the digest using the template below.
  7. Display the digest and optionally export as Markdown.

Digest Template

# Monthly Digest — [Month Year]

## Key Numbers
| Metric | Amount | vs. Last Month |
|--------|--------|----------------|
| Total Income | $X,XXX | +/-XX% |
| Total Expenses | $X,XXX | +/-XX% |
| Net (Income - Expenses) | $X,XXX | — |
| Savings Rate | XX% | +/-X pts |

## Spending by Category
| Category | Amount | % of Total | vs. Last Month |
|----------|--------|------------|----------------|
| ... | ... | ... | ... |

## Top 5 Expenses
1. [Description] — $XXX on [Date]
2. ...

## Anomalies & Alerts
- [flag icon] [Description of anomaly]

## New This Month
- First time seeing: [vendor name]

Without Wilson

To create a monthly digest manually:

  1. Export the month's transactions from your bank as CSV.
  2. Open in Google Sheets and add a Category column if missing.
  3. Key numbers:
    Total Income:   =SUMIFS(Amount, Amount, ">0", Date, ">="&DATE(2025,3,1), Date, "<="&DATE(2025,3,31))
    Total Expenses: =ABS(SUMIFS(Amount, Amount, "<0", Date, ">="&DATE(2025,3,1), Date, "<="&DATE(2025,3,31)))
    Savings Rate:   =(Income - Expenses) / Income * 100
    
  4. Category breakdown: Create a pivot table grouped by Category, summing Amount.
  5. Top expenses: Sort by Amount ascending, take the first 5 rows.
  6. Month-over-month: Keep last month's totals in a separate tab and calculate =(ThisMonth - LastMonth) / LastMonth * 100 for each category.
  7. Anomaly detection (manual): Calculate the average and standard deviation per category over the last 6 months. Flag any category where this month's spending exceeds the average + 2 standard deviations.

Important Notes

  • The digest defaults to the previous calendar month. Ask for a specific month if needed.
  • Savings rate is calculated as (income - expenses) / income. If income is zero, savings rate is shown as N/A.
  • Anomaly detection uses a rolling 6-month window. New users with less history will see fewer anomaly flags.
  • The digest does not include transfers between your own accounts (e.g., savings transfers) in the income/expense totals.

Comments

Markdown supported