Controlled document rule check
Invoice check: a €30 total mismatch
The local rule returned review for the €30 mismatch, pass for the matching control and unavailable when the second page was omitted.
What this check covers
We ran the existing document calculation rules on deliberately prepared observations from fictional text specimens. This isolates the arithmetic and its page-coverage guard, so the result is attributable to that rule rather than to a model’s reading of a file.
- Material
- Fictional plain-text specimen; no customer or bank data
- Input preparation
- Fields, quotes, currency and page coverage specified manually
- Executed
- Local evaluateDocumentCalculation function; three variants
- Outside this check
- OCR, PDF/image extraction, live API, billing and AI detection
- Evidence
- Original specimen text, structured inputs, recorded outputs and SHA-256 hashes
The question
Will the invoice-total rule flag a total that differs from the supplied subtotal and tax, while refusing to complete the check with missing-page coverage?
How the controlled check was set up
The discrepancy variant contains an intentionally inconsistent stated amount. The matching control changes only that amount. The incomplete variant declares two pages while supplying observations from page one only. All amounts are explicit EUR values, and every observation cites its specimen line. These confidence labels were supplied for the controlled input; they are not measured extraction confidence.
The supplied amounts
€350.00 subtotal + €70.00 tax = €420.00; the discrepancy specimen states €450.00.
Specimen text · English
FICTIONAL CONTROLLED SPECIMEN - NOT A REAL FINANCIAL DOCUMENT Invoice calculation example Variant: discrepancy Page 1 of 1. All specimen amounts are listed below. Currency: EUR Subtotal: 350.00 EUR Tax amount: 70.00 EUR Total: 450.00 EUR No real customer, issuer, bank account, tax identifier or transaction is represented. These amounts are supplied as structured observations; OCR and file extraction are not exercised.
Expected and observed results
| Variant | Page coverage | Expected status | Observed status | Calculated / stated |
|---|---|---|---|---|
| Intentional discrepancy | 1 of 1 pages | review | review | 420.00 EUR / 450.00 EUR |
| Matching control | 1 of 1 pages | pass | pass | 420.00 EUR / 420.00 EUR |
| Missing second page | 1 of 2 pages; page 2 omitted | unavailable | unavailable | — |
review means the supplied figures disagree; pass means this calculation agrees; unavailable means the rule cannot reach a result within the supplied scope. None of these statuses verifies the document’s authenticity.
What the finding means
The stated total exceeds the supplied components by €30.00. This case checks addition of supplied totals only. It does not validate tax rates, item quantities, a supplier, delivery, payment or a legal invoice. A genuine invoice may explain a difference with an additional charge or discount that is missing from the available evidence.
What to do with a similar real document
- Compare the subtotal, tax, total and currency against the original invoice and confirm that all pages are available.
- Look for shipping, other fees, discounts, credits and explicit rounding adjustments before treating a mismatch as unexplained.
- Request a corrected invoice or supporting explanation. For payment reconciliation, compare the invoice with supplied statement entries in the separate Cross-check workflow.
Inspect the evidence
The English specimen text and JSON files preserve the exact input and output of this local run. The files are test evidence, not live API reports or documents to submit as genuine records.
Execution and reproducibility
Recalculate the displayed amounts and compare the expected result with the recorded output. The downloads include every operand, citation and coverage flag. SHA-256 hashes identify the exact bytes used. Our local regression runner verifies the same artifacts against the recorded version of the calculation function; this does not reproduce file extraction or a complete paid analysis.
- Executed at (UTC)
- · Node v24.19.0
- Rules version
document-rules-2026-09-25-v1- Calculation module SHA-256
aaeae93eeec726bbb384c408aea35146450c2d49a25e80308c152291bec3f650- Input JSON SHA-256
d4439d6cc77e840ab38c89fc7e7ffaf3e4002931e1ef48e1f41d88cfa802aabc- Result JSON SHA-256
39309b28f8e7c61dc909e872a03a351cb747f10eef3689575add2e08ede47dfd
Limits of this evidence
This is a small, deliberately constructed rule check. It measures neither real-world fraud detection nor OCR quality, image/video accuracy, false-positive rates or customer outcomes. Two cases and six variants do not constitute the public detector benchmark and do not substantiate the product-reported 98.8% figure.