# Brave Search (MCP Server)

> Web, local, news, image and video search via the Brave Search API

- Type: MCP server
- Publisher: Brave (official)
- Categories: Research
- Website: https://github.com/brave/brave-search-mcp-server
- Source: https://github.com/brave/brave-search-mcp-server
- Transport: stdio
- Launch command: npx -y @brave/brave-search-mcp-server --transport stdio
- Sign-in: api-key
- Environment variables: BRAVE_API_KEY
- GitHub stars: 1479
- Listed: 2025-06-12

## About

Brave's official MCP server connects the Brave Search API to AI assistants with web, local, image, video, news and place search, AI summarization and LLM-ready context retrieval. It runs locally via npx or Docker and requires a Brave Search API key.

## Install

1. **Add the server to Codex**

   Codex CLI:

   ```bash
   codex mcp add brave-search --env BRAVE_API_KEY=YOUR_BRAVE_API_KEY -- npx -y @brave/brave-search-mcp-server --transport stdio
   ```

   config.toml:

   ```toml
   [mcp_servers.brave-search]
   command = "npx"
   args = ["-y", "@brave/brave-search-mcp-server", "--transport", "stdio"]
   
   [mcp_servers.brave-search.env]
   BRAVE_API_KEY = "YOUR_BRAVE_API_KEY"
   ```

   Agents SDK:

   ```python
   from agents import Agent, Runner
   from agents.mcp import MCPServerStdio
   
   async with MCPServerStdio(
       name="Brave Search",
       params={"command": "npx", "args": ["-y", "@brave/brave-search-mcp-server", "--transport", "stdio"]},
   ) as server:
       agent = Agent(name="Assistant", mcp_servers=[server])
       result = await Runner.run(agent, "Search the web for the best open-source vector databases")
       print(result.final_output)
   ```

## Tools

- `brave_web_search`
- `brave_local_search`
- `brave_place_search`
- `brave_news_search`
- `brave_image_search`
- `brave_video_search`
- `brave_summarizer`
- `brave_llm_context`

## Example prompts

- "Search the web for the best open-source vector databases"
- "Find coffee shops near Union Square open now"
- "Get the latest news about SpaceX launches"
- "Summarize search results on Rust async runtimes"

## Related

- [Fetch](https://chatgptplugins.directory/mcp/fetch.md): Fetch web pages and convert them to markdown for LLMs
- [DeepWiki](https://chatgptplugins.directory/mcp/deepwiki.md): Ask questions about any public GitHub repo using DeepWiki's generated docs
- [Apify](https://chatgptplugins.directory/mcp/apify.md): Run thousands of Apify Actors for scraping and automation from your agent
- [Firecrawl](https://chatgptplugins.directory/mcp/firecrawl.md): Scrape, search, crawl and interact with live web pages for LLMs
- [AWS Knowledge](https://chatgptplugins.directory/mcp/aws-knowledge.md): Search current AWS docs, API references and regional availability
- [Exa](https://chatgptplugins.directory/mcp/exa.md): AI-native web search, page fetching and multi-step research via Exa

---
Source: https://chatgptplugins.directory/mcp/brave-search · Index: https://chatgptplugins.directory/llms.txt · Independent and community-run. Not affiliated with or endorsed by OpenAI. ChatGPT is a trademark of OpenAI.