This lets Claude search the web through Serper API when you need current information to pair with Nansen's on-chain data. You can run multiple queries in parallel, get up to 20 results per query, and it filters out paywalled sites like Bloomberg and Twitter automatically. Returns clean JSON with titles, links, snippets, and sometimes knowledge graph data. It's meant for grabbing fresh context like token prices, news, or ecosystem updates before diving into the blockchain analytics. If you need full page content from specific URLs, there's a separate fetch command for that.
npx skills add https://github.com/nansen-ai/nansen-cli --skill nansen-web-searcher