ocr

Extracts text from images, photos, scans, screenshots, and scanned PDFs using PaddleOCR. Returns line-level text with optional bounding box coordinates. Strong accuracy for CJK, small print, and handwritten text.

Captures current OS clipboard (text/image) for analysis or action without pasting into chat. Supports Korean and English triggers.

Analyzes media files (PDFs, images, diagrams, screenshots) using a vision backend to extract structured data, descriptions, or summaries instead of literal file reading.