Brave Search MCP in Claude Code

Quick answer: Install the Brave Search MCP server with npx -y @modelcontextprotocol/server-brave-search, add the JSON block below to ~/.claude/settings.json, restart Claude Code, and run /mcp to confirm the connection. Setup runs about 5 minutes on a fresh machine, verified on @modelcontextprotocol/server-brave-search as of April 15, 2026.

The Brave Search MCP server gives Claude Code a web search tool backed by the Brave Search API. The model can run a query, read snippets, and pull URLs directly into its working context without you pasting search results by hand. The server is published as @modelcontextprotocol/server-brave-search and hits the Brave web and local search endpoints under the hood.

This guide covers what you get after the wiring is done, the exact config, verification steps, prompt patterns that tend to work well, and the 4 issues that trip people up most often in the first week.

What you get when it is connected

Once the Brave Search server is attached, Claude Code can call the server tools from inside any conversation. You do not invoke the tools by hand. When you ask Claude a question the model decides which tool to call and parses the response for you. For teams that live inside Brave Search day to day, this replaces dozens of context switches per week with a single line in chat.

Two tools ship in the server: brave_web_search and brave_local_search. Each returns a structured JSON with title, URL, and snippet. There is no document fetch tool, so for full-page content Claude will chain into a separate fetch server (mcp-server-fetch works well alongside).

Prerequisites

A Brave Search API key from api.search.brave.com. The free Data for AI tier gives 2000 queries per month, enough for day-to-day research inside Claude Code. Node 20 or later on the host.

If you use a version manager like nvm or asdf for Node, confirm the version Claude Code inherits. Open a terminal, run node -v, and note the output. Claude Code uses the Node it sees on PATH at launch, so a shell profile that sets the right version is the reliable path.

Install via npx

Run the package once with npx to verify it starts cleanly:

npx -y @modelcontextprotocol/server-brave-search

The first run downloads the package (a few MB) and starts the server on stdio. The server does not print much on success - it waits for MCP protocol messages on stdin. Press Ctrl-C to stop it. The actual runtime setup happens through Claude Code itself in the next step.

If the install fails with a network error, your npm registry may be blocked. Set npm config set registry https://registry.npmjs.org and retry. Behind a corporate proxy, also set HTTP_PROXY and HTTPS_PROXY in your shell.

Add the config block to ~/.claude/settings.json

Open ~/.claude/settings.json in your editor. If the file does not exist yet, create it with {} as the starting content. Add an mcpServers object with an entry for this server:

{
  "mcpServers": {
    "brave-search": {
      "command": "npx",
      "args": ["-y", "@modelcontextprotocol/server-brave-search"],
      "env": {
        "BRAVE_API_KEY": "BSA_XXX"
      }
    }
  }
}

Save the file. If you already have other MCP servers defined, merge the new entry into the existing mcpServers object rather than replacing it.

Restart Claude Code fully (quit and reopen, not just close the window). The server is spawned lazily on the first tool call in a session, not at launch, but the config is read once per Claude Code start.

Verify the connection

Open a new Claude Code session and type /mcp at the prompt. You should see the server listed with a green or connected indicator. If it shows as failed, click into it for the stderr output - the error message usually points at the problem directly (bad token, wrong path, missing Node).

Run a trivial first prompt to confirm round trips work. Good smoke tests:

• For read servers: ask for a list of whatever resource type it exposes.
• For write servers: ask for a describe on a known resource first, then try a safe write on a test resource.

If the first prompt works, the wiring is done. From here on you interact with the server purely through normal prompts in Claude Code.

Example prompts that work well

Here are prompts that tend to get good responses once the server is attached:

• Search the web for the Next.js 16 async params migration guide and give me the top three results with a one-line summary each.
• Find recent articles about GPU spot pricing on AWS from the last 30 days and list the sources.
• Look up the current signature of the Stripe charges API and paste the relevant docs URL.
• Run a local search for coffee shops near Kreuzberg Berlin and show me the top five with ratings.
• Search for security advisories published this week that mention Postgres 16, and tell me which ones apply to our setup.
• Find the GitHub release notes for React 19 and give me the three biggest user-facing changes.

Claude will chain tool calls on its own when the prompt implies several steps. For a summarize-then-write flow the model will often call read tools first, then a single write tool at the end. If a prompt keeps burning tool calls, narrow it: specify the resource ID, the time range, or the exact field you want rather than asking Claude to scan everything.

Environment variable security

The API key goes in BRAVE_API_KEY. Keep it out of shared config files. On macOS and Linux, export it in ~/.zshrc or ~/.bashrc and reference the variable from your Claude config. On Windows, set it in the user environment variables panel rather than inline in the JSON. Rotating the key is cheap: regenerate at the dashboard and restart Claude Code.

A general rule across every MCP server: never paste secrets directly into settings.json that lives in a shared or git-tracked directory. Keep the actual secret values in your shell profile (~/.zshrc, ~/.bashrc, or a 1Password-cli helper), export them at shell start, and reference the variable names from the Claude config. That way the secret stays on your machine and the config file is safe to share with teammates.

On macOS, terminals launched from Spotlight or from the Dock both inherit the shell profile. If you launch Claude Code from a GUI shortcut that does not go through a shell, env vars may not propagate - launch from a terminal instead.

Troubleshooting

Searches return 401. The API key is wrong or has not been activated yet. New keys sometimes take a couple of minutes to propagate. Check the key at api.search.brave.com/app/keys and confirm the status is Active.

Searches hit 429 rate limit. You crossed the monthly quota on the Data for AI tier. Upgrade to a higher tier or wait until the next billing window. The dashboard shows usage against the cap.

Local search returns empty results. Local search coverage is thinner than web search. Try the web search tool with a location qualifier in the query, or supply latitude and longitude if the prompt allows.

Results look stale. Brave caches high-traffic pages for roughly 24 hours. Add freshness=pd to the query (past day) by asking Claude to include a recency filter.

For any issue not listed here, the first step is /mcp inside Claude Code to see the current status and any recent stderr from the server. The second step is running the exact npx command from your terminal to see if the server starts cleanly outside Claude Code. Between those two checks, most problems become obvious within a minute.

Next steps

Once the Brave Search server is attached and verified, the useful next move is writing a short prompt template you keep in your notes. List the 3 or 4 prompts you run most often against this server, and paste them into Claude Code when needed. Over a few weeks you build a personal command library that gets real work done without typing much.

For team projects, commit a .mcp.json at the repo root with the same structure. Everyone on the team gets the server wired up automatically on first open, and individual secrets stay in shell profiles. That is the setup pattern that scales past a single developer.