🔍 Prysm MCP Server

What is Prysm MCP Server?

Prysm MCP Server is a Model Context Protocol server that enables AI assistants like Claude to scrape web content with high accuracy and flexibility. It provides multiple scraping modes, format outputs, and image support.

How to use Prysm MCP Server?

Install globally via npm (npm install -g @pinkpixel/prysm-mcp or prysm-mcp), then configure an MCP client with the command npx -y @pinkpixel/prysm-mcp. Use tools such as scrapeFocused, scrapeBalanced, scrapeDeep, and formatResult to scrape and format web content. Output directories can be customized via environment variables or tool parameters.

Key features of Prysm MCP Server

• Multiple scraping modes: focused (speed), balanced (default), deep (thorough)
• Content analysis to determine the best scraping approach
• Format results as markdown, HTML, or JSON
• Optional image extraction and local download
• Configurable scroll behavior for single-page applications
• Save formatted results to custom output directories

Use cases of Prysm MCP Server

• AI assistants scraping news articles or blog posts for summarization
• Extracting product data from e-commerce sites for analysis
• Gathering documentation or technical content for reference
• Archiving web pages with images into structured files

FAQ from Prysm MCP Server

What dependencies or runtime does Prysm MCP Server require?

It runs on Node.js and uses Puppeteer for web scraping. Installation is via npm.

How do I configure the output directory for scraped files?

Set the PRYSM_OUTPUT_DIR and PRYSM_IMAGE_OUTPUT_DIR environment variables, or specify output paths directly in tool parameters. Default is ~/prysm-mcp/output/.

What transport protocol does Prysm MCP Server use?

It uses the Model Context Protocol (MCP) over standard I/O, designed for integration into MCP-compatible clients like Claude Desktop, Cursor, and Windsurf.

How do I run the server in debug mode?

Use the command DEBUG=mcp:* node bin/prysm-mcp to enable debug logging.