Screenshot
Captures and tiles web page screenshots into AI-friendly dimensions, specifically optimized for Claude Vision API.
This tool efficiently captures high-quality screenshots of web pages, uniquely designed for AI vision workflows. It automatically adjusts resolutions and tiles full pages into 1072x1072 pixel chunks, ensuring optimal processing compatibility with models like Claude Vision API. By integrating seamlessly with Model Context Protocol (MCP) clients, it provides fresh, up-to-date visual content for AI analysis without caching.
主要功能
01Claude Vision API optimized with automatic 1072x1072 resolution limiting
02Always fresh content with no caching
03Fast screenshot capture using Puppeteer
041 GitHub stars
05Seamless MCP integration for AI workflows
06Automatic tiling of full pages into AI-friendly chunks
使用案例
01Providing visual context to AI models for analysis and understanding
02Automating screenshot capture for AI vision workflows
03Testing website responsiveness across configurable viewports