Expense Receipt Organizer

Automatically OCR and organize travel receipts, invoices, payment screenshots into a neat reimbursement package.

When to Use

• User needs to报销 (company reimbursement)
• Have multiple PDF invoices + image payment screenshots
• Want auto OCR recognition + summary table + packaged zip

Core Steps

1. Collect files — scan folder for PDF invoices and image screenshots
2. OCR recognition — use tesseract.js to extract text from images/PDF
3. Information extraction — extract amount, date, route, payee from recognized text
4. Generate CSV — create formatted summary table with bank info
5. Package files — put summary + all original files into one folder
6. Zip it — compress to a single zip file ready to send to company
7. Clean up — move temporary files to trash

Dependencies

• Node.js/npm installed (comes with OpenClaw)
• No extra binary required (tesseract.js runs in Node, no system tesseract needed)

Usage

User: Help me organize my reimbursement receipts
Assistant: 
1. Ask user where the files are
2. Ask for reimbursement person name + bank account info
3. Run the process:
   - npm install tesseract.js in temp folder
   - OCR all images
   - extract information
   - generate CSV
   - package everything
   - clean up
4. Tell user where the final zip is located

Example Output

~/Desktop/{Name}_reimbursement_{Date}.zip
├── {Name}_reimbursement.csv  (summary table with bank info)
├── *.pdf (original invoices)
└── *.jpg (original payment screenshots)

Notes

• Uses CSV format for maximum compatibility (opens in Excel/WPS/Numbers)
• Keeps original files attached for company audit
• Cleans up temporary files after packaging
• Currently optimized for train ticket reimbursement, easily extended to other receipt types