Exa Search

Add to your client

Copy the config for your MCP client and paste it into its config file.

{
  "mcpServers": {
    "exa-search": {
      "command": "npx",
      "args": [
        "-y",
        "exa-mcp-server"
      ],
      "env": {
        "EXA_API_KEY": "<your-exa-api-key>"
      }
    }
  }
}

Step-by-step guides: Add to Claude Desktop · Add to Cursor · Add to Windsurf

Before you start

• Node.js with npx available (for the local stdio install)
• An Exa API key from the Exa dashboard at https://dashboard.exa.ai/api-keys (set as EXA_API_KEY)
• Alternatively, no local runtime is needed if you use the hosted remote endpoint at https://mcp.exa.ai/mcp

About Exa Search

Tools & capabilities (6)

When to use it

• Use it when you want your AI agent to pull clean, parsed article text instead of messy HTML or thin snippets.
• Use it when you need neural/semantic search that finds conceptually relevant pages rather than keyword matches.
• Use it when you already have a URL and want the model to read the full page content (web_fetch_exa).
• Use it when you need precise filtering by domain, date range, or content type via the advanced search tool.
• Use it when you want a zero-install hosted MCP endpoint to prototype quickly before self-hosting.

Quick setup

1. 1Get an Exa API key from https://dashboard.exa.ai/api-keys.
2. 2For local use, add the server to your MCP client config with command `npx -y exa-mcp-server` and env `EXA_API_KEY`.
3. 3Alternatively, configure the hosted remote endpoint `https://mcp.exa.ai/mcp` (no local runtime required).
4. 4Restart your MCP client (Claude Desktop, Cursor, VS Code, etc.) so it loads the server.
5. 5Verify by asking the assistant to run a web search and confirm web_search_exa returns results.

Exa Search FAQ