image-ocr
Installation
SKILL.md
When to Use
- You need to extract text from an image file (PNG/JPEG/TIFF/BMP) for downstream processing or review.
- You want to run OCR with a specific Tesseract language model (e.g.,
eng,chi_sim). - You prefer providing a natural-language request that contains an image path (e.g., "Interpret the image at ...") instead of manually setting
image_path. - You need a quick local OCR verification workflow from the command line.
- You want a simple JSON-configured OCR runner that can be integrated into scripts or automation.
Key Features
- OCR text extraction using Tesseract via
pytesseract. - Supports common image formats: PNG, JPEG, TIFF, BMP (via Pillow).
- Multi-language OCR through the
langconfiguration option. - Natural-language request parsing to automatically locate the image path.
- Config-driven execution through
scripts/ocr_config.json.