tencentcloud-ocr-vatinvoice_skill

This skill converts invoice images into structured VAT data by calling Tencent OCR for full-field extraction of VAT invoices, including PDFs.
  • Python

2.5k

GitHub Stars

2

Bundled Files

2 months ago

Catalog Refreshed

3 months ago

First Indexed

Readme & install

Copy the install command, review bundled files from the catalogue, and read any extended description pulled from the listing source.

Installation

Preview and clipboard use veilstrat where the catalogue uses aiagentskills.

npx veilstrat add skill openclaw/skills --skill tencentcloud-ocr-vatinvoice

  • _meta.json314 B
  • SKILL.md9.3 KB

Overview

This skill calls Tencent Cloud VatInvoiceOCR to extract full-field data from VAT and electronic invoices. It returns structured fields such as invoice codes, numbers, dates, amounts, buyer/seller info, line items and supports PDF input. Use it when you need reliable, end-to-end invoice OCR and structured output for downstream processing.

How this skill works

The skill accepts an image or PDF (URL or Base64) and invokes Tencent Cloud OCR advanced invoice API. It first classifies the invoice type, routes to the appropriate recognition engine (including full-electric invoices), then maps detected fields into a standardized JSON structure with header, parties, amounts, personnel and itemized lines. PDF pages are supported when IsPdf is enabled and PdfPageNumber is set.

When to use it

  • Extract all invoice fields including codes, dates, amounts and tax details
  • Recognize VAT special, VAT ordinary, electronic special/ordinary and plain electronic invoices
  • Parse PDF invoices or specific PDF pages into structured data
  • Extract itemized line entries (service/product list) for accounting or automation
  • Batch-process invoices for AP automation or expense reconciliation

Best practices

  • Provide ImageUrl when possible for faster, more stable processing (store images in Tencent COS if available)
  • Ensure input images/PDFs are under 10MB after Base64 and between 20–10000 pixels
  • Enable IsPdf explicitly for PDF files and set PdfPageNumber >=1 for multi-page PDFs
  • Prefer high-contrast, legible scans to reduce ImageBlur and decoding errors
  • Handle error codes gracefully (download, decode, quota, billing, or service-not-enabled errors)

Example use cases

  • Automate accounts payable: extract invoice header, tax, and line items into AP system
  • Expense auditing: validate invoice codes, dates, and tax amounts against records
  • ERP data entry: convert batches of invoices (images or PDFs) into structured entries
  • Compliance reporting: aggregate tax rates, totals and buyer/seller tax IDs for reporting

FAQ

PNG, JPG, JPEG and PDF. GIF is not supported. For PDF enable IsPdf and set PdfPageNumber if needed.

What fields are returned?

Full invoice headers, buyer/seller name and tax IDs, amounts, tax rates, personnel, detailed line items and optional polygon coordinates for each field.

What are common error causes?

File download failures, empty or blurred images, size limits (>10MB after Base64), decode failures, service not enabled, quota or billing issues.

Built by
VeilStrat
AI signals for GTM teams
© 2026 VeilStrat. All rights reserved.All systems operational