Analyze visual inputs via an MCP server with stdio transport.
npx skills add https://github.com/Dwsy/zai-vision-skill --skill zai-vision
Please help me install this Agent Skill. Skill: zai-vision Source: https://github.com/Dwsy/zai-vision-skill/tree/main Command: npx skills add https://github.com/Dwsy/zai-vision-skill --skill zai-vision
This Skill bridges the gap between visual information (screenshots, diagrams, videos) and actionable digital content like code, text, and analysis, streamlining development and understanding workflows.
Use the zai-vision skill to convert the UI screenshot at '/path/to/ui-screenshot.png' into React code.
High-intent search queries and answers about installing and using this skill.
To convert a UI screenshot into code, this Skill analyzes the visual input and generates structured code, prompts, or design specifications. It uses an MCP server to process the image and output the corresponding code implementation directly.
Yes, you can extract text from any image or screenshot using OCR capabilities. The Skill processes the visual input through its MCP server to identify and output the embedded text, streamlining data entry and documentation workflows.
To analyze technical diagrams like architecture and flowcharts, the Skill interprets the visual structure and translates it into structured data or textual descriptions. This allows you to understand complex visual logic and integrate it into your development context.
Yes, an MCP server is required to run visual analysis. The Skill leverages a Model Context Protocol server with stdio transport to dynamically load tools and execute visual processing tasks while optimizing context window usage.
The best way to diagnose errors from a screenshot is to use the Skill's visual analysis to interpret the error message. It analyzes the visual input and suggests actionable fixes, bridging visual information and troubleshooting workflows.
Yes, you can process video files for insights. The Skill provides multimodal video analysis capabilities, allowing you to extract structured data, textual descriptions, and actionable insights from visual video inputs.
CyapstayeIntegrate external Model Context Protocol servers with stdio and HTTP transport.
LuminaVaultExtract text from WhatsApp images with Tesseract OCR and save to Obsidian.
PolGubauCreate, distribute, and manage custom Expo development clients for iOS and Android.
JamkrisTranslate visa document images into bilingual PDFs with OCR and rotation normalization.
