Three-way match purchase orders against delivery and invoice docs with ChatGPT agent
- Posted
- by Justin (@jand__)
- Last checked
- Method:
- How we verify
Evidence: Screenshots or logsThe source shows screenshots, a transcript or the agent's output.
Outcome
In about 4 minutes ChatGPT agent checked that all 5 POs had matching delivery and invoice documents with no variances, combined the PDFs, and wrote a short report.
Poster says:Done in around 4 minutes
This is the source's own number. We haven't measured it.
Agent
Prompt Recreated prompt
The source didn't share its exact prompt, so we wrote this from what it describes.
Here's a task I'd like your help with. Read it all the way through before you start. Goal: Check that every purchase order in my Google Drive folder matches its delivery document and its invoice, and flag anything that doesn't line up. Check with me first: which folder should you use, and is there anything specific I want you to look for, like quantity, price or missing paperwork? What I'll give you: - Access to the Google Drive folder - The purchase orders, delivery documents and invoices in that folder Rules: - Match each purchase order to its delivery document, then to its invoice. - Keep a simple .md checklist as you go: one line per order, what it matched, and whether anything was off. - If something doesn't match, note it next to that order in the checklist. - If you can't tell which document goes with which order, or one is missing, ask me instead of guessing. - Don't change, move or delete my original files. Output: 1. One combined PDF per order (order + delivery document + invoice) 2. A short report saying which orders matched and which had problems 3. The finished checklist file Stop there and hand everything back to me to review. You'll know it worked when: every order is marked either "matches" or "problem found," with details for any problems.
Try this in:
Sends the Recreated prompt shown above. ChatGPT and Grok open with it filled in, and ChatGPT may send it straight away. Gemini opens empty, so we copy the prompt for you to paste. Every button also copies it.
Opens in a new tab. We don't see what you send.
These links are only a shortcut. We're not affiliated with OpenAI, Google or xAI, and your results may not match the source.
Optional stronger version
Our rewrite with extra checks. The poster didn't test it.
Here's a task I'd like your help with. Read it all the way through before you start. Goal: Three-way match each purchase order with its delivery note and invoice in my Drive and flag mismatches. Quick check first: tell me if anything below is missing or unclear. Inputs (fill in or confirm): - Drive folder link - How files are named or linked (PO number) - Tolerance for price or quantity differences - Which currency and tax treatment invoices use - My PDF naming pattern for the combined files Rules: - Match documents by PO number; list any document without a match. - Compare line items: quantity, unit price, totals, dates. - Flag differences above the tolerance with exact values. - Don't edit or move any files. - Treat partial deliveries as their own case: match delivered quantity to what was invoiced. - Summarize totals: matched, mismatched, missing documents, and the money at stake. - Keep a running .md checklist with one line per PO as you go. - Combine each PO, delivery note and invoice into one PDF named by the pattern. - If one document could belong to two POs, ask instead of guessing. If you get stuck: If a PDF is a scan you can't read, mark it 'needs review'. If a login, CAPTCHA or two-step code appears, stop and ask me to handle it; never guess or store passwords. Verify before you finish: Re-check flagged items by reopening the documents. Output: A spreadsheet: PO, delivery, invoice, status, differences. Done when: every PO is matched or flagged with a clear reason.
Source
https://x.com/jand__/status/1946283011445784991
Related cases
- Have Claude Cowork prepare late-invoice reminder drafts every Monday night, with a report of who owes whatRecreated promptEvidence: Described by posterBeginner-friendly
Michel, in San Diego, works in a tight-margin business with basic back-office software, and sending late-invoice reminders was repetitive manual work. Claude Cowork walked them through setting up an agent that logs in every Monday night, checks overdue accounts receivable and matches customer names against a Google Sheet of contacts. Using Michel's email template, it adds the contact name, invoice number, amount and due date, puts the emails in the Drafts folder and sends Michel a report of which customers were included, what each owes and the totals. Michel reviews the drafts and sends them, saving about an hour every Monday night.
Claude Cowork · Ops · Source date:
- Find a list of missing invoices across work and personal email with ChatGPT agentRecreated promptEvidence: Described by poster
Testing ChatGPT agent, a poster gave it a list of invoices he owed his operations assistant, and he says it found all of them in his personal and work email. He also notes it booked accommodation successfully (they had to contact the hotel to cancel the test booking) and estimates about a 40% end-to-end completion rate on complex tasks for now.
ChatGPT agent · Ops · Source date:
- Split, rename and sort about 200 scanned invoices with Claude CoworkRecreated promptEvidence: Described by poster
After the poster scanned almost 200 invoices, Claude Cowork split them out, renamed them, sorted them into folders and created a summary Excel file.
Claude Cowork · Ops · Source date: