What are the prerequisites?

You need Node.js 18 or higher, a Kagi API key (currently in closed beta – request an invite from support@kagi.com), and access to the search API.

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.

What credentials does it need?

It reads `KAGI_API_KEY` from the environment. Store these in your client's env block or a secrets manager — never commit them to a repository — and grant the narrowest scope that still covers your use case.

SSyncDev

Start a project

Kagi (Node.Js) MCP Server

Node implementation of a Kagi MCP server

Local serverstdioTypeScript

What is the Kagi (Node.Js) MCP Server?

Node implementation of a Kagi MCP server

The short version

A Node.js implementation of the Kagi MCP server that provides web search capabilities to AI assistants using the Kagi search API.

Web search using the Kagi API
Support for multiple search queries in parallel
Formatted search results
Built with TypeScript and the official MCP SDK
Pre-built and ready to use (no build step required)
Unique tool name (kagi_web_search) to avoid conflicts with other MCP servers

What it needs from you

Node.js 18 or higher - Kagi API key

More MCP options

Compare MCP servers by category, or browse hosted MCP servers if you prefer an endpoint over a local process.

How to install the Kagi (Node.Js) MCP Server

{
  "mcpServers": {
    "kagi": {
      "command": "npx",
      "args": ["-y", "github"],
      "env": {
        "KAGI_API_KEY": "your-value"
      }
    }
  }
}

Add to claude_desktop_config.json, then restart Claude Desktop.