A modern MCP (Model Context Protocol) server implementation that provides AI assistants with web search capabilities via the Brave Search API and
Connect Brave Gemini Research to Claude, Cursor or any other MCP client and it stops being a tab you switch to. A modern MCP (Model Context Protocol) server implementation that provides AI assistants with web search capabilities via the Brave Search API and advanced research paper analysis with Google's Gemini model. The brave gemini research mcp server is what makes that connection.
This project enables AI assistants like Claude to perform web searches and analyze research papers directly through a standardized API interface. The MCP server exposes three main tools:
The server ships on npm as tsc, so your MCP client can launch it on demand — there is no separate build step. Add the server block to your client's configuration, restart it, and the tools register themselves.
The toolset is worth reading before you wire it up, because it tells you what the integration is really for:
Windows — C:\Users<username>\AppData\Roaming\Claude\claude_desktop_config.jsonmacOS — ~/Library/Application Support/Claude/claude_desktop_config.jsonFunction — brave_web_searchParameters — - query (required): Search query (max 400 chars)Options — "summary", "critique", "literature review", "key findings", "comprehensive" (default)Summary — Comprehensive overview including research question, methodology, key findings, and conclusionsCritique — Critical evaluation of methodology, validity, limitations, and suggestions for improvementComprehensive — Complete analysis covering all aspects (default)Tools — Functions with defined schemasTransports — Communication channels between clients and serversHandlers — Logic to process requests and return responsesPrerequisites — The Prerequisites tool exposed by this serverConfiguration is passed through the environment: BRAVE_API_KEY, GOOGLE_API_KEY, DEBUG. Treat anything key-shaped as a real credential — scope it to the minimum the server needs, and rotate it if it ever lands in a shared config.
This sits in the developer tooling group, where several servers overlap in what they claim to do but differ sharply once you actually set them up. Brave Gemini Research's toolset — Windows, macOS, Function and 11 more — is a fair guide to whether it matches your workflow. It is maintained by falahgs; worth a glance at recent repository activity before you build anything load-bearing on it.
This entry was verified against Brave Gemini Research's own documentation before publication; SyncDev keeps the directory reviewed rather than auto-generated.
| Tool | What it does |
|---|---|
| Windows | C:\Users\<username>\AppData\Roaming\Claude\claude_desktop_config.json |
| macOS | ~/Library/Application Support/Claude/claude_desktop_config.json |
| Function | brave_web_search |
| Parameters | - query (required): Search query (max 400 chars) |
| Options | "summary", "critique", "literature review", "key findings", "comprehensive" (default) |
| Summary | Comprehensive overview including research question, methodology, key findings, and conclusions |
| Critique | Critical evaluation of methodology, validity, limitations, and suggestions for improvement |
| Comprehensive | Complete analysis covering all aspects (default) |
| Tools | Functions with defined schemas |
| Transports | Communication channels between clients and servers |
| Handlers | Logic to process requests and return responses |
| Prerequisites | The Prerequisites tool exposed by this server. |
| Installation | 1. Clone the repository: bash git clone https://github.com/falahgs/brave-gemini-research-mcp.git cd brave-gemini-research-mcp |
| Building | The Building tool exposed by this server. |
{
"mcpServers": {
"brave-gemini-research": {
"command": "npx",
"args": ["-y", "tsc"],
"env": {
"BRAVE_API_KEY": "your-value",
"GOOGLE_API_KEY": "your-value",
"DEBUG": "your-value"
}
}
}
}Add to claude_desktop_config.json, then restart Claude Desktop.
| Variable | Description | Required |
|---|---|---|
| BRAVE_API_KEY | Credential the server authenticates with. | Yes |
| GOOGLE_API_KEY | Credential the server authenticates with. | Yes |
| DEBUG | Configuration value read at startup. | Optional |
The simplest web tool that matters — fetch any URL and get model-ready markdown back.
Industrial-strength web extraction — render, scrape, crawl and search entire sites into clean markdown.
Puppeteer-powered browser control that drives pages from the accessibility tree instead of pixels.
Search built for AI, not humans — semantic web search that returns model-ready content, plus code context.
Answers, not links — delegate questions to Perplexity's search-grounded models and get cited responses back.
Put 6,000+ pre-built scrapers at your assistant's fingertips through one MCP endpoint.