Brave Search MCP is a server tool by Mikechao, built with TypeScript/Node.js. It connects to the Brave Search API, acting as an MCP server for Search. Distributed as an npm package and Docker image, it offers various search tools for web, local, image, video, and news content. This free program enables AI agents to access current information, supporting tasks like market research and LLM grounding. Its design supports multiple transport protocols, making it adaptable for various AI client integrations.

What does Brave Search MCP connect to?

The Brave Search MCP server, developed by Mikechao, connects directly to the Brave Search API. This program functions as a Model Context Protocol (MCP) server specifically for Search, enabling AI agents to interact with Brave's search capabilities. It is implemented using TypeScript/Node.js and is available as an npm package or a Docker image. This design allows it to serve as a critical bridge between AI clients and the vast information accessible through web searches.

Core features and tools

Brave Search MCP exposes several callable tools. These include brave_web_search for general web queries with advanced filtering, and brave_local_search for finding local businesses. For media, it provides brave_image_search with metadata, and brave_video_search for web videos. News access is handled by brave_news_search. Additionally, it features brave_summarizer for AI-powered summaries and brave_llm_context to retrieve web content optimized for AI agents and RAG pipelines.

Who is it for?

Brave Search MCP targets developers building AI agents or MCP-compatible tools, AI engineers, and LLM developers. It provides essential search functions for enhancing AI capabilities. This program is suitable for those needing real-time web data to improve agent responses and support advanced AI workflows. Skip this if you do not require a Brave Search API Key for your AI agent or development environment.

How to get started

The server authenticates using an API key passed as an environment variable, such as BRAVE_API_KEY, and supports STDIO, HTTP, and SSE transport protocols.

Use cases

• Enabling AI assistants to perform real-time web searches to access current information, reducing hallucinations and providing more informed responses.
• Automating market research and business intelligence by allowing agents to run searches across multiple angles and return structured reports with sources.
• Facilitating LLM grounding and RAG pipelines by retrieving pre-extracted web content optimized for AI agents to ground answers in fresh, relevant web content.

Brave Search MCP offers a specialized suite of search tools for AI agents, grounding LLMs in current web data. Its support for various AI clients and transport protocols makes it adaptable for diverse development environments. Mikechao's free program addresses the need for accurate, timely information in AI systems, helping to reduce AI hallucinations. Start by integrating brave_web_search into an AI assistant to fetch current information.