Overview
What is Kagi MCP Server (Node.js)?
A Node.js implementation of the Kagi MCP server that provides web search capabilities to AI assistants using the Kagi search API. It is built with TypeScript and the official MCP SDK, and is pre‑built for immediate use.
How to use Kagi MCP Server (Node.js)?
Requires Node.js 18+ and a Kagi API key (currently in closed beta – request an invite from [email protected]). Run directly with npx github:elliottlawson/kagi-mcp-server or clone the repo and run node build/index.js. Configure in Claude Desktop, Cursor, or Claude Code using the provided MCP configurations.
Key features of Kagi MCP Server (Node.js)
- Web search using the Kagi API
- Pre‑built, no build step required
- Unique tool name (
kagi_web_search) avoids conflicts - Built with TypeScript and the official MCP SDK
- Supports multiple search queries in parallel
- Formatted search results
Use cases of Kagi MCP Server (Node.js)
- AI assistants performing web searches for up‑to‑date information
- Integration with Claude Desktop, Cursor, and Claude Code
- Avoiding tool name conflicts with other MCP search servers
- Retrieving Kagi search results directly in AI conversations
FAQ from Kagi MCP Server (Node.js)
What is the Kagi MCP Server (Node.js)?
It is a Node.js implementation of an MCP server that provides web search capabilities to AI assistants using the Kagi search API.
What are the prerequisites?
You need Node.js 18 or higher, a Kagi API key (currently in closed beta – request an invite from [email protected]), and access to the search API.
How do I set it up?
You can run it directly with npx: KAGI_API_KEY=your_key npx github:elliottlawson/kagi-mcp-server. Alternatively, clone the repo and run node build/index.js. No build step is required.
How does it avoid conflicts with other MCP servers?
The tool is named kagi_web_search instead of a generic search, preventing conflicts with other MCP search servers.
What transport does it use?
The server uses stdio for communication with the MCP client.