eject-dream

zai-vision

0
0
# Install this skill:
npx skills add eject-dream/glm-mcp-skills --skill "zai-vision"

Install specific skill from multi-skill repository

# Description

AI image/video analysis - OCR, UI-to-code, error diagnosis, diagrams, charts, and comparisons.

# SKILL.md


name: zai-vision
description: AI image/video analysis - OCR, UI-to-code, error diagnosis, diagrams, charts, and comparisons.


ZAI Vision

Analyze images and videos with AI.

Commands

Command Usage Purpose
ui2code python scripts/zai_vision.py ui2code <img> <type> "<prompt>" UI → code/prompt/spec/description
ocr python scripts/zai_vision.py ocr <img> "<prompt>" [lang] Extract text
error python scripts/zai_vision.py error <img> "<prompt>" Diagnose errors
diagram python scripts/zai_vision.py diagram <img> "<prompt>" Analyze flowcharts/architecture
dataviz python scripts/zai_vision.py dataviz <img> "<prompt>" Analyze charts/graphs
diff python scripts/zai_vision.py diff <before> <after> "<prompt>" Compare two UIs
image python scripts/zai_vision.py image <img> "<prompt>" General image analysis
video python scripts/zai_vision.py video <video> "<prompt>" Analyze video (≤8MB)

Examples

python scripts/zai_vision.py ui2code screenshot.png code "Generate React component"
python scripts/zai_vision.py ocr terminal.png "Extract the error message"
python scripts/zai_vision.py diagram arch.png "Explain this architecture"
python scripts/zai_vision.py diff design.png impl.png "List differences"

When to Use

Scenario Command
Convert UI design to code ui2code
Extract text from screenshot ocr
Debug from terminal screenshot error
Understand flowchart/architecture diagram
Analyze dashboard/chart dataviz
Compare design vs implementation diff
General image question image
Video content analysis video

# Supported AI Coding Agents

This skill is compatible with the SKILL.md standard and works with all major AI coding agents:

Learn more about the SKILL.md standard and how to use these skills with your preferred AI coding agent.