Side-by-side comparison of two Model Context Protocol servers — install paths, tools, quality signals, and directory engagement so you can pick the right one for Claude, Cursor, and other MCP clients.
Work with Kagi without API access (you'll need to be a customer, tho). Searches and summarizes. Uses Kagi session token for easy authentication.
Crawl websites into clean Markdown, search pages, and extract structured data with LLMs. Built-in MCP server for web research and RAG pipelines.
Quality signal
60/100 (Good)
51/100 (Fair)
Install path
npx · high
npx · high
Engagement
3 0 0 41
5 0 1 3
Tools
kagi_search_fetchkagi_summarizer
Crawl single pages or entire websitesStrip navigation, scripts, and boilerplate from HTMLGenerate clean Markdown files with citationsProduce structured JSONL index of crawled pagesLocal embedding model included, no API keys requiredConfigurable URL filters and download of referenced files