What it does

Brave's own server, and the one to use now that the older reference implementation sits in the archived repository. Returns web, news, image, and local results as structured data rather than a paragraph. That makes it the better choice when you want the agent to cite sources or compare pages, and the worse choice when you just want a summary.

Install

Terminal

npx -y @brave/brave-search-mcp-server

Client configuration

Paste this into your MCP client config (Claude Desktop, Cursor, or any client that reads mcpServers), then restart the client.

mcp.json

{
  "mcpServers": {
    "brave-search-official": {
      "command": "npx",
      "args": [
        "-y",
        "@brave/brave-search-mcp-server"
      ]
    }
  }
}

Use it for

Search that returns real result lists with URLs you can then fetch

Skip it if

Answering questions directly; this returns sources, not summaries

When to use it

Reach for this when the task needs search inside the agent loop. Skip it if a plain shell command or a one-off script is enough — every server you add costs context and adds a failure mode.

Other TypeScript servers

Filesystem — Gives an agent scoped read, write, and search access to directories you explicitly allow
Memory — A persistent knowledge graph stored in a local JSON file, exposed through entities, relations, and observations
Sequential Thinking — Gives the model a scratchpad for structured, revisable reasoning: numbered thoughts that can branch, revise, and conclude

Brave Search (official)