agentsclimarketplace
Sign in

Invoice extractor

Skill Uhudsavasindankacanokcu2/finance-skills-for-claude/invoice-extractor

Extract structured data from invoices, receipts, and bills (pasted text, PDF, or image), then categorize for bookkeeping and flag anomalies. Produces clean line-item data ready for accounting import plus duplicate/overcharge/tax warnings. Use when the user shares invoices, receipts, bills, or asks to process/categorize expenses or prepare data for their accountant.From its SKILL.md

Install
npx -y skills add Uhudsavasindankacanokcu2/finance-skills-for-claude --skill invoice-extractor

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

2 things to look at

  • no licenseNo license file was found in the repository. Code published without one is not open source by default, so using it at work is a question for whoever answers licensing questions where you are.
  • 1 stars1 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

SKILL.md

2.8 KB, 582 tokens by cl100k_base, as published. Nobody here has run it

Invoice & Expense Extractor

You act as a meticulous bookkeeper. Convert any invoice/receipt into clean, categorized, import-ready data and catch the errors that cost money.

When to use

Invoice/receipt paste, PDF, or image; "categorize these expenses", "prep for accountant", "what did we spend on X", expense report, VAT/tax extraction.

Extract these fields per document

  • vendor (name), vendor_tax_id (if present)
  • invoice_number, invoice_date, due_date
  • currency
  • line_items: [{description, qty, unit_price, line_total}]
  • subtotal, tax_rate, tax_amount, total
  • payment_method / paid_status if shown
  • expense_category (see categories below)

Categorization

Assign each invoice ONE primary category: Software/SaaS, Payroll/Contractors, Rent/Utilities, Marketing/Ads, Travel, Equipment/Hardware, Professional Services, Supplies, Taxes/Fees, Other. If genuinely ambiguous, set REVIEW — never guess on money.

Validation & warnings (the value)

Run these checks and report failures with the exact numbers:

  • Math check: does sum(line_totals) + tax == total? If not, flag the discrepancy and the delta.
  • Tax check: does subtotal * tax_rate ≈ tax_amount? Flag mismatches (common overcharge / wrong-rate error).
  • Duplicate check: same vendor + same amount + date within 7 days = possible double-billing → flag.
  • Round-number / estimate flag: suspiciously round totals on a "final" invoice.
  • Missing-field flag: no invoice number, no tax ID where required, no date.
  • Currency mismatch: line items in different currency than total.

Output

  1. Clean table of all extracted invoices (one row each, all key fields) — formatted for copy-paste into a spreadsheet/accounting tool.
  2. CSV block with the same data (headers: vendor,invoice_number,date,due_date,currency,subtotal,tax,total,category) so it imports directly.
  3. Category summary: total spend per category for the batch.
  4. Warnings: numbered, each with the document, the issue, and the $ impact.
  5. Action list: which invoices to dispute/recheck and why.

Rules

  • Extract only what's actually in the document. Mark unreadable/missing fields as null, never fabricate a number or tax ID.
  • For images/scans, note any low-confidence reads explicitly so the user verifies them.
  • Keep amounts at full precision; preserve the document's currency.
  • Note that final tax treatment should be confirmed with their accountant.

What ships with it

Read from the repository

Just SKILL.md. No reference files, no scripts.

Gives 0 of the 12 instructions most finance skills give in 582 tokens

Counted across 469 of the 469 authors here whose files we hold, read 2026-08-07

  • Extract date vendor amount and descriptionin 15 of 469, across 3 files
  • Scan folder for invoice filesin 14 of 469, across 2 files
  • Rename files to standard formatin 14 of 469, across 2 files
  • Show organization plan before movingin 14 of 469, across 2 files
  • Generate summary CSVin 14 of 469, across 2 files
  • Organize files by categoryin 13 of 469, across 1 file
  • Preserve original filesin 13 of 469, across 1 file
  • Flag files missing critical infoin 13 of 469, across 1 file
  • Produce the requested output filein 9 of 469, across 4 files
  • Build best, base, and worst case scenariosin 9 of 469, across 5 files
  • Implement backoff if rate limit errors occurin 8 of 469, across 3 files
  • Determine the weighted average cost of capitalin 8 of 469, across 4 files

Said here and by no other author read

  • extract listed fields from each document
  • assign each invoice one primary category
  • use REVIEW for ambiguous categories
  • run the math check on totals
  • run the tax rate check
  • flag suspicious round-number totals

Grouped from the skills themselves: near-identical wordings counted once, and counted by distinct author, so one author publishing three of these counts once. Length counted with cl100k_base; the agent that loads this file may tokenize it differently.

Keep looking

Skills are one crate of 325,949. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.