Daily Scan

Overview

Turn phone photos of documents into searchable PDFs with OCR and stable filenames based on capture date and headline text. Preserve the original photo, generate a readable scan-like PDF, and support later retrieval of saved scan files.

Runtime Requirements

• Default OCR path: local Tesseract CLI
• Required local dependencies for the stable path:
  • tesseract
  • Python packages used by bundled scripts: opencv-python or cv2, Pillow, and either reportlab or ocrmypdf depending on the active PDF path
• Optional/experimental OCR path:
  • PaddleOCR-based script exists but is not the default stable engine
• No cloud upload is required for core operation
• The skill assumes bundled helper scripts under scripts/ are present and callable by the host agent

Workflow