skills.rest
sungaoxiang-backend
sungaoxiang-backendCOMMUNITY
Explore all skills by this author →
GitHub

evidence-invoice-handler

Extract structured card_features from VAT invoices via OCR slot filling.

Updated Jul 20, 2025
Search Tags:#compliance#evidence#OCR#invoice#vat-invoice#card_features
One-click install
npx skills add https://github.com/sungaoxiang-backend/intelligent-evidence-platform --skill evidence-invoice-handler
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: evidence-invoice-handler
Source: https://github.com/sungaoxiang-backend/intelligent-evidence-platform/tree/main/app/agentic/skills/evidence-invoice-handler
Command: npx skills add https://github.com/sungaoxiang-backend/intelligent-evidence-platform --skill evidence-invoice-handler

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill automates the extraction of key evidence slots from VAT invoices by converting them into a standardized card_features list, reducing manual data entry and increasing reliability.

Core Features & Use Cases

  • Rule-based extraction: captures required slots such as type, code, number, supplier and buyer details, amounts, and dates.
  • Enhanced extraction: enables additional analysis like business relationship insights and tax burden considerations.
  • Use Case: Upload an invoice image and receive a ready-to-use card_features payload for evidence cards in investigations.

Quick Start

To extract features from an invoice image, provide a YAML payload including evidence_materials with the image URL and identifiers, and specify whether to use the default rule-based mode or the enhanced mode.

Frequently Asked Questions about evidence-invoice-handler

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I extract structured evidence data from VAT invoices?▼

You can extract structured evidence data from VAT invoices by using OCR-based slot filling to capture required fields like code, number, amounts, and dates, outputting a flat card_features list for evidence analysis.

What specific fields can I capture from VAT invoices for compliance checks?▼

You can capture specific fields including invoice type, code, number, supplier and buyer details, amounts, and dates using the default rule-based extraction mode for compliance evidence.

Can I analyze business relationships and tax burdens from invoice images?▼

Yes, you can analyze business relationships and tax burdens from invoice images by enabling the enhanced extraction mode, which provides deeper insights beyond standard rule-based slot filling.

Do I need external data sources to process VAT invoice OCR extraction?▼

No, you do not need external data sources to process VAT invoice OCR extraction; the skill requires only predefined inputs and slot names to generate the structured evidence card payload.

How do I format the input payload to parse VAT invoices into evidence cards?▼

To parse VAT invoices into evidence cards, provide a YAML payload containing evidence_materials with the invoice image URL and identifiers, then specify whether to use default or enhanced mode.

What is the difference between default and enhanced extraction modes for invoice processing?▼

The difference is that default mode captures standard slots like amounts and dates, while enhanced mode adds deeper analysis into business relationships and taxation burden insights for invoice processing.

Related Skills

Explore More Legal & Compliance Skills→
sungaoxiang-backendsungaoxiang-backend

evidence-household-handler

Extract household registry image data into structured card_features slots.

Community
Advanced
sungaoxiang-backendsungaoxiang-backend

evidence-alipay-transfer-handler

Extract structured card_features from Alipay transfer screenshots.

Community
Advanced
legoutlegout
3

image-ocr

Extracts text and structured data from images using OCR engines.

Community
Advanced
sungaoxiang-backendsungaoxiang-backend

evidence-license-handler

Extract structured card_features from business license images via OCR.

Community
Advanced