What problem does it solve? Whole-image previews of very tall, wide, high-resolution, or text-dense images get downscaled, making text unreadable and causing vision models to hallucinate content. This Skill inspects image dimensions, deterministically slices abnormal images into overlapping tiles with a reading-order manifest, and guides ordered extraction with overlap reconciliation so no content is invented or lost. ## Core Features & Use Cases - Dimension Inspection: Reads width, height, aspect ratio, and orientation from the original file before deciding whether tiling is needed. - Deterministic Tiling: Slices images in vertical, horizontal, or grid mode with configurable overlap, producing PNG tiles plus a manifest.json with reading order. - Overlap Reconciliation: Instructs how to merge tile extractions, deduplicate boundary content, and flag unreadable regions instead of guessing. - Use Case: A user shares a long scrolling chat screenshot whose preview text is illegible. The Skill detects the extreme aspect ratio, generates top-to-bottom tiles, and extracts the full conversation in order without inventing messages. ## Quick Start Use the read-special-images skill to extract all text from this long screenshot without guessing any unreadable parts.
