agentleFS
aashariaashari/ai-agent-skills/skills/apple-mail/mail-expenses/SKILL.md

mail-expenses

Skill7 starsChanged 9 months agoMITView on GitHub

Extract financial transactions, expenses, receipts, payments, and invoices from email. Summarize spending with amounts, merchants, and categories. Use when user asks about expenses, spending, receipts, payments, or financial transactions from email. Arguments: time range like "last 24 hours", "this month", "last week", or a specific date range.

The file

We split this skill into the rules it gives an agent

1 rule. About 0 are common, 1 is distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈1

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Use SQLite date functions — never compute Unix epochs manually (wrong year risk).

    line 20

≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Mail Expenses — Financial Transaction Summary

Period: $ARGUMENTS (default: last 24 hours)

Step 1: Find financial emails

Use SQLite date functions — never compute Unix epochs manually (wrong year risk).

DB="$HOME/Library/Mail/V10/MailData/Envelope Index"

# Date range from $ARGUMENTS. Use SQLite date expressions directly:
# "yesterday"   → dt >= date('now','-1 day','localtime') AND dt < date('now','localtime')
# "today"       → dt >= date('now','localtime')
# "last 7 days" → dt >= date('now','-7 days','localtime')
# "this month"  → dt >= date('now','start of month','localtime')
# Always print the range before querying so the period is visible in output.

sqlite3 "$DB" "
SELECT datetime(m.date_received,'unixepoch','localtime') as dt,
       s.subject, a.address as sender, mb.url, m.ROWID
FROM messages m
JOIN subjects  s  ON m.subject = s.ROWID
JOIN addresses a  ON m.sender  = a.ROWID
JOIN mailboxes mb ON m.mailbox = mb.ROWID
WHERE datetime(m.date_received,'unixepoch','localtime') >= date('now','-1 day','localtime')
  AND datetime(m.date_received,'unixepoch','localtime') <  date('now','localtime')
  AND m.deleted = 0
  AND mb.url NOT LIKE '%Spam%' AND mb.url NOT LIKE '%Trash%'
  AND mb.url NOT LIKE '%Sent%'
  AND (
    s.subject LIKE '%receipt%'
    OR s.subject LIKE '%invoice%'
    OR s.subject LIKE '%payment%'
    OR s.subject LIKE '%order%'
    OR s.subject LIKE '%purchase%'
    OR s.subject LIKE '%charge%'
    OR s.subject LIKE '%transaction%'
    OR s.subject LIKE '%billing%'
    OR s.subject LIKE '%renewal%'
    OR s.subject LIKE '%subscription%'
    OR s.subject LIKE '%paid%'
    OR s.subject LIKE '%refund%'
    OR s.subject LIKE '%transfer%'
    OR s.subject LIKE '%successful%'
    OR s.subject LIKE '%confirmation%'
    -- Indonesian / multilingual
    OR s.subject LIKE '%bukti%'
    OR s.subject LIKE '%pembayaran%'
    OR s.subject LIKE '%tagihan%'
    OR s.subject LIKE '%transaksi%'
    OR s.subject LIKE '%berhasil%'
    -- Common financial senders
    OR a.address LIKE '%bank%'
    OR a.address LIKE '%paypal%'
    OR a.address LIKE '%stripe%'
    OR a.address LIKE '%xendit%'
    OR a.address LIKE '%livin%'
    OR a.address LIKE '%gopay%'
    OR a.address LIKE '%ovo%'
    OR a.address LIKE '%dana%'
    OR a.address LIKE '%apple%'
    OR a.address LIKE '%google%'
    OR a.address LIKE '%amazon%'
    OR a.address LIKE '%shopee%'
    OR a.address LIKE '%tokopedia%'
    OR a.address LIKE '%grab%'
  )
ORDER BY m.date_received DESC;" 2>/dev/null

Step 2: Parse amounts from email bodies

python3 ~/.claude/skills/_mail-shared/parser.py <ROWID1> <ROWID2> ...

Step 3: Extract amounts with Python regex

From each parsed body, extract monetary amounts:

import re

patterns = [
    r'(?:Rp|IDR)\s*[\d.,]+',          # Indonesian Rupiah
    r'(?:USD|US\$|\$)\s*[\d.,]+',      # USD
    r'(?:SGD|S\$)\s*[\d.,]+',          # Singapore dollar
    r'Total[:\s]+(?:Rp|IDR|USD|\$|€|£)[\d.,]+',
    r'Amount[:\s]+(?:Rp|IDR|USD|\$|€|£)[\d.,]+',
    r'(?:€|£|¥|₹|RM|THB|PHP|VND)\s*[\d.,]+',
]

Step 4: Categorize transactions

  • Transfers (bank to bank, BI-Fast, QRIS, virtual account)
  • Bills/Subscriptions (recurring charges, renewals)
  • Shopping (e-commerce, marketplace)
  • Food/Lifestyle (restaurants, delivery, groceries)
  • Travel (flights, hotels, transport)
  • Refunds (money coming back)
  • Credit Card Payments (paying off balances — not new spending)

Output Format

Financial Activity — [PERIOD]

Line-by-line per account (no markdown tables — WhatsApp/Telegram render them as raw text):

💳 [Bank/Wallet Name]

HH:MM · [Merchant] · Rp X.XXX

Subtotals by category. Total: Rp X.XXX

Note refunds separately. Flag duplicates and unusually large transactions proactively.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.