This MCP server, developed by kazuph, enables web content fetching and image processing capabilities for AI assistants. It integrates with libraries like jsdom, Readability, and Sharp to extract readable content from web pages, convert HTML to Markdown, and process images for optimal viewing. The implementation handles robots.txt restrictions, merges multiple images vertically with size limits, and uses macOS clipboard operations for seamless image insertion. It's designed for scenarios requiring web scraping, content summarization, or image handling within AI conversations, allowing assistants to fetch and process web content on demand while respecting web standards and optimizing for readability.
Fetches web content from a specified URL. Parameters: url (string, required), maxLength (integer, default: 20000), startIndex (integer, default: 0), raw (boolean, default: false), enableFetchImages (boolean, default: false), ignoreRobotsTxt (boolean, default: false).
Extracts and displays the title of the article from the fetched web content.
Processes images from web pages. Parameters: imageMaxCount (integer, default: 3), imageQuality (integer, default: 80), imageMaxHeight (integer, default: 4000), imageMaxWidth (integer, default: 1000), enableFetchImages (boolean, default: false).
Supports pagination for both text and images. Parameters: imageStartIndex (integer, default: 0).
Automatically optimizes images as JPEG for better performance.
Extracts the first frame from animated GIFs.
No reviews yet. Be the first to review!
Sign in to join the conversation
Highlight AI