Optical character recognition and image analysis prompts for extracting, analyzing, and processing visual content. These prompts handle various image types from screenshots to scientific documents with high precision.
- image-analyzer.md - Image Analyzer: Generates structured descriptions of images including objects, text, context, and visual elements
- ocr-assistant.md - OCR Assistant: High-precision conversion of scientific and technical images to GitHub Flavored Markdown
- ocr-translator.md - OCR Translator: Language detection and translation of OCR-extracted text with formatting preservation
Most prompts suggest:
temperature: 0.2 # Low creativity for accurate text extraction and analysis
reasoning_effort: "high" # Careful handling of OCR errors and complex visual elements