evo-invoice-fraud-detector
SkillCommerce & financeImplements fraud detection pipeline with fuzzy vendor matching via rapidfuzz, priority-ordered fraud checks, and JSON report generation.
Available today. Use it from your connected AI after setup.
No other account needed.
Connect ahel once, and every AI you use reads what you have installed.
Then ask your AI: use the evo-invoice-fraud-detector skill
What this skill tells your AI
The instructions your AI receives, as published by openlair/openskill in tasks-evolved/invoice-fraud-detection/environment/skills/evo-invoice-fraud-detector/SKILL.md and read by ahel’s review.
Detects invoice fraud using 5 priority-ordered rules.
Fraud Rules (Priority Order)
- Unknown Vendor - vendor name not in vendors database (fuzzy match score < 85)
- IBAN Mismatch - vendor exists but IBAN doesn't match
- Invalid PO - PO number doesn't exist in PO database
- Amount Mismatch - PO exists but amount differs by > 0.01
- Vendor Mismatch - PO valid but linked to different vendor ID
Key Functions
find_vendor_match(vendor_name, vendor_dict)- Fuzzy match using token_set_ratio, cutoff=85check_invoice_fraud(invoice, vendor_dict, po_dict)- Check single invoicerun_fraud_detection_pipeline(invoices, vendor_dict, po_dict)- Check all invoicessave_fraud_report(flagged, output_path)- Save JSON report
Usage
import sys
sys.path.insert(0, '/app/environment/skills/evo-pdf-invoice-parser/scripts')
sys.path.insert(0, '/app/environment/skills/evo-reference-data-loader/scripts')
sys.path.insert(0, '/app/environment/skills/evo-invoice-fraud-detector/scripts')
from utils import extract_all_invoices
from utils import load_vendor_database, load_po_database # name collision - use aliases
# Better approach: import modules directly
import importlib
parser = importlib.import_module('utils') # depends on sys.path order
Signals
- GitHub stars
- 89
- Forks
- 4
- Last commit
- Sep 2026
Advanced
- Catalog kind
- skill
- Gateway key
evo-invoice-fraud-detector- Source
- github.com/openlair/openskill