NCCN Guidelines

This MCP server provides AI assistants with access to the National Comprehensive Cancer Network (NCCN) clinical practice guidelines through automated web scraping, PDF downloading, and content extraction capabilities. Built using Python with httpx for async HTTP requests, BeautifulSoup for HTML parsing, and pdfplumber for PDF text extraction, the implementation automatically scrapes NCCN's website to build a searchable index of guidelines, handles login authentication for PDF downloads with intelligent caching, and extracts specific pages from guideline PDFs while preserving layout and table structures. The server maintains a 7-day cache for both the guidelines index and downloaded PDFs to minimize server load, supports targeted page extraction for efficient content retrieval, and includes comprehensive logging for debugging, serving healthcare professionals and AI assistants requiring programmatic access to evidence-based cancer treatment protocols and clinical decision support without manual navigation of the NCCN website.