invoice-processing
Skill1 starsChanged 7 days agoMITView on GitHub
Extract invoice data (PDF, image, text) into structured JSON, validate totals, NIP and IBAN, and export to a spreadsheet. Use when processing, checking or booking invoices or receipts.
In this repository7 other files it gives its agents.
- angular-conventionsSkill
- appstore-payouts-to-excelSkill
- clean-git-summarySkill
- conventional-commitSkill
- create-commitSkill
- ksef-invoiceSkill
- xlsx-toolkitSkill
The file
We split this skill into the rules it gives an agent
2 rules. About 0 are common, 2 are distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈2
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
- Distinctive: line 8
Read each document (PDF text, image, email).
- Distinctive: line 33
Never pay, send or upload anything.
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 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.
Files
Invoice Processing
- Read each document (PDF text, image, email). One invoice per JSON object, in this shape:
{"number":"FV/12/2026","type":"invoice","issue_date":"2026-11-03","sale_date":"2026-11-03",
"due_date":"2026-11-17","currency":"PLN",
"seller":{"name":"Acme sp. z o.o.","tax_id":"1234560007","country":"PL"},
"buyer":{"name":"Jan Kowalski IT","tax_id":"PL9876540016","country":"PL"},
"lines":[{"description":"Hosting 11/2026","qty":1,"unit_net":100.00,"net":100.00,"vat_rate":"23","vat":23.00}],
"totals":{"net":100.00,"vat":23.00,"gross":123.00},
"iban":"PL61109010140000071219812874","category":null,"notes":[]}type:invoice|correction|proforma|receipt. Corrections keep negative amounts.vat_rate: a number as text, orzw/np/oo(exempt, not taxable, reverse charge).- Copy values exactly as printed. Never guess a missing field — set
nulland add a note. - Dates
YYYY-MM-DD, amounts as numbers with a dot.
- Run
python scripts/validate_invoice.py invoices.json(path relative to this skill). It checks line math, totals, gross = net + VAT, Polish NIP checksum, IBAN checksum, due date, and duplicates (same seller + number). Fix extraction mistakes and re-run; issues that are really on the document stay as notes for the user — never "fix" the invoice's numbers. - Category: use the user's list or chart of accounts if given; otherwise propose short categories (e.g. software, hosting, hardware, travel, marketing, services) and mark them as proposals.
- Export: one row per invoice (or per line if asked) to CSV or to .xlsx with the
xlsx-toolkitskill — columns: number, issue_date, due_date, seller, seller_tax_id, net, vat, gross, currency, category, status, notes.
Hard rules:
- Foreign-currency invoices: keep the original currency. Convert only if the user asks, and then state the rate, its date and source.
- Never pay, send or upload anything. Tax treatment is a hint for the accountant, not advice.
- Report: count processed, count with issues, and each issue in one line.
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
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.

