Liu-Zhangzhu

1 mods across 1 repository, 2 stars between them.

vision-reader

01

Liu-Zhangzhu/opencode-vision-skill

Skill Claude CodeCodex

Use when the user asks to read, view, or analyze a PDF, image (PNG/JPG/GIF/BMP/TIFF/WEBP), PowerPoint (PPT/PPTX), or take a screenshot. Provides visual description (understands layout, arrows, boxes, diagrams), OCR text extraction, PDF structure extraction, and PPT content extraction. Triggered by keywords: "read this…

2 1mo ago A 132 tokens original MIT