Invoice OCR Recognition and Archiving
Paste the following prompt into your AI chat to install this skill:
Please follow https://skillhub.cn/install/skillhub.md to install @user_4cdf8f25/invoice-ocr-v1.
About this skill
The Problem
Manual invoice processing usually breaks down after inspection: information is scattered across images, PDF files, and archives, while invoice code, number, tax ID, amount, tax rate, and trip details must be transcribed into a spreadsheet. Railway tickets and tax-issued invoices add non-standard fields. After extraction, teams still need to classify rows, check duplicates, flag items for validation, and produce a summary for finance or reimbursement.
How It Works
The skill accepts invoice images, PDF files, or archives, decompresses inputs when needed, and processes each file into a structured Excel workbook with Invoice Details, Summary Statistics, and Exceptions and Attention sheets:
- Field extraction: captures 29 fields, including basic invoice data, buyer and seller details, amount, tax rate, expense category, accounting subject, trip details, and validation results.
- Invoice types: supports VAT special invoices, ordinary invoices, electronic ordinary invoices, passenger transport e-invoices, railway e-tickets, and tax-issued invoices, using a prioritized type detector.
- Text fallback: prefers
PDFtext layers or OCR, degrades gracefully when Chinese language data is missing, and can useRapidOCRas a Chinese image OCR path without relying onTesseract chi_sim. - Validation: scores buyer tax ID, seller tax ID, invoice number, tax rate, and issue date; suspicious rows are highlighted, while identical invoice numbers are flagged as possible duplicates.
- Summarization: aggregates amounts and counts using custom classification rules, producing KPIs, category tables, and a pie chart for reimbursement or cost review.
Boundaries
This skill is best for invoice cleanup, pre-reimbursement checks, and batch tabulation. It does not replace online tax validation; invoices above the threshold or with lower validation scores still need manual review. Blurred scans, missing fields, and non-invoice files may be marked for manual entry or cleared to avoid junk data, making it useful as an initial screening and archiving workflow rather than a final accounting authority.
Use Cases
- Before month-end reimbursement, extract passenger tickets, railway tickets, and ordinary invoices from an archive into Excel and flag duplicates or threshold breaches.
- While reviewing travel invoices, extract PDF and scan data into 29 fields, then summarize by expense category and generate a pie chart.
- Before entering the reimbursement system, batch convert VAT special invoices, ordinary invoices, and railway tickets into a detail sheet and mark suspicious rows.
Best For
- Finance reviewers who need a batch of invoice images or PDFs turned into a checkable line-item sheet.
- Operations managers who need passenger tickets, train tickets, and ordinary invoices grouped by expense category with anomalies surfaced.
- Accountants who need tax IDs, amounts, rates, and subjects extracted with threshold and duplicate flags.
Related Skills
Organizes files by extension into subfolders like Documents, Code, and Archives, then outputs a report.
Extract tables, formulas, charts, and layout from invoices, reports, papers, and multi-column documents.
Generates a multi-sheet Excel report containing only structured data tables from byteplan-analysis results.
Automatically sort directory files into type-based folders, with dry-run preview, reports, and JSON custom rules.