Analyzes web content and adds it to a knowledge base for question answering.
Memex is a Model Context Protocol (MCP) tool inspired by Vannevar Bush's Memex project, designed to analyze web content and build a knowledge base for enhanced question answering. It leverages APIs from Claude, FireCrawl, and Voyage to process web content referenced in user queries. The resulting knowledge base is stored as Markdown files, ideally viewed with Obsidian, allowing users to ask questions about the analyzed content and receive informed answers based on the collected information.
Key Features
01Web content analysis
02Knowledge base creation
03Integration with Claude, FireCrawl, and Voyage APIs
04Markdown-based storage for knowledge base
05Designed for use with Obsidian
Use Cases
01Answering questions based on web content
02Building a knowledge base from web sources
03Integrating web content analysis into question answering workflows