When should you use it?
- Payment notices clutter your inbox
- You want a weekly view without a finance app
- You use n8n
Not for: Accounting records you need exact; Shared mailboxes
What do you need, and what do you get?
You need
- Mail label for payment notices
- n8n with mail and model credentials
You get
- Weekly digest email with payments and stats
How does it work, step by step?
- Label the noticesRoute payment emails to one label or filter.
- Fetch the weekPull the past seven days from that label on a schedule.
- Combine and trimAggregate the messages and cut to a safe length for the model.
- Write the digestPrompt for one line per payment and a short stats block.
- Email itSend to yourself every Sunday.
The prompt
The original prompt by Marianna Nakos, word for word, with a link to where it was published. Below it is our adapted version with fill-in blanks, if you want it. Works in n8n and ChatGPT. Values in lime are placeholders.
Original prompt by Marianna Nakos
Hi, here's your payments digest for this week: ## Payment Summary - Week of [Date] - [Vendor] - $[Amount] - [Date] - [Continue for each payment] ## Weekly Stats: - Total Spent: $[Calculate sum] - Payment Count: [Number] - Largest Payment: [Vendor and amount] Thanks, Your AI agent
Exactly as published by Marianna Nakos in N8n Tutorial: Built an AI Workflow Agent to summarize insights on labeled emails on payment notifications.
Our adapted version (optional)
Below are this week's payment notification emails. Write a digest with one line per payment: vendor, amount, currency and date. Then add totals: amount spent, number of payments and the largest payment. If an email is not a payment, skip it. If an amount is unclear, list it under needs checking. Emails: [text]
What can go wrong?
- Too many emails at once can exceed model limits; the author truncates the combined text.
- Check totals against your bank now and then; summaries can drop a line.
Our take
A tidy beginner workflow, and the truncation node is a practical detail many tutorials skip. Turning the prompt into a fixed template keeps the digest stable. Routing every payment notice to one label first is what keeps the flow simple; the n8n job then only has to fetch a week from one place. Marianna's truncation step guards against sending more text than the model can take, which is the failure most beginner flows hit. Keep the prompt fixed: one line per payment and a short stats block make weeks easy to compare. Treat the digest as a view, not a ledger. Summaries can drop a line, so compare the totals against your bank statement now and then, and never use the digest as an accounting record. Read-only mail access is enough for the whole job. If a vendor floods the label, unsubscribe or filter at the mail level rather than asking the model to ignore it.
Where the time goes back: A week of scattered notices becomes a two-minute read.
FAQ
Why truncate?
What stats are included?
Source and credit
The original prompt on this page is by Marianna Nakos, shown word for word with a link to where it was published. The summary, notes and adapted version are ours.
- Original
- N8n Tutorial: Built an AI Workflow Agent to summarize insights on labeled emails on payment notifications
- Creator
- Marianna Nakos
- Published on
- mariannanakos.substack.com (Substack)
- Original date
- Aug 16, 2025
- License
- Not openly licensed; the original prompt is shown word for word with credit and a link to the original
- Originally for
- n8n
- AI dry run
- Oct 23, 2026: passed. Dry-run tested by our AI test agent on sample input. PASS: Gates held on desk check; no live sends or purchases.
AI dry run on sample data; not run with live connected accounts. The playbooks.bot test agent filled the prompt with synthetic inputs, produced the output and checked it against the page's stated outputs and guardrails. - Verified
- Oct 24, 2026: source link resolves to the original, the credited creator published it, our excerpt or prompt matches it, and the license allows our use.
Are you the creator? Claim or correct this page.