Skip to main content
Glama
mcp2everything

MCP2Brave

by mcp2everything
OverviewSchemaSourceRelated ServersDiscussionsAdmin
Python
Remote
README.md
# mcp2brave

一个基于MCP协议的服务器，使用Brave API实现网络搜索功能。

## 系统要求

- Python 3.11+
- UV包管理器
- Brave API密钥

## 安装步骤

1. 克隆项目
```bash
git clone <仓库地址>
cd mcp2brave
```

2. 创建并编辑环境变量文件
```bash
# 添加Brave API密钥到.env文件
.env
```

3. 使用UV创建虚拟环境
```bash
# 创建并激活虚拟环境
uv venv
# Windows系统使用:
.venv\Scripts\activate
# Linux/Mac系统使用:
source .venv/bin/activate
```

4. 安装依赖
```bash
uv sync
```

## 使用方法

### 安装为Claude扩展
```bash
fastmcp install mcp2brave.py
```

### 开发模式与MCP检查器
要使用MCP检查器测试功能：

```bash
fastmcp dev mcp2brave.py
```

运行后，可以在浏览器访问MCP检查器：http://localhost:5173

## 可用工具

- `search_web(query: str)`: 使用Brave API搜索网络
- `search_web_info(query: str)`: 同上，带中文描述

## 环境变量

- `Brave_API_KEY`: 你的Brave API密钥（必需）

## 注意事项

- 确保在使用前已正确设置API密钥
- 虚拟环境激活后才能运行相关命令
- 如遇到编码问题，请确保系统使用UTF-8编码

## 手动添加Cline Continue Claude
打开Cline Continue Claude的MCP服务器配置文件，加入以下信息
```json

"mcp2brave": {
      "command": "uv",
      "args": [
        "run",
        "--with",
        "fastmcp",
        "--with",
        "python-dotenv",
        "--with",
        "beautifulsoup4",
        "--with",
        "requests",
        "fastmcp",
        "run",
        "C:\\Users\\你的真实路径\\mcp2brave.py"
      ],
      "env": {
        "BRAVE_API_KEY": "API密钥"
      }
    }
```


## 详细操作步骤
### Cline集成
![配置](./images/1.png)

### Cline集成
![配置](./images/2.png)
### Cline集成
![配置](./images/3.png)

### 使用示例
![搜索示例](./images/dev.png)
Deploy Server
One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

TDQS

C2.6/5.0

Scored across 6 tools

Disambiguation2/5

Multiple tools have unclear boundaries and overlapping purposes. 'brave_search_summary' and 'search_brave_with_summary' appear to serve the same function with only minor wording differences. Similarly, 'get_url_content_direct' and 'url_content' are essentially duplicates with one in English and one in Chinese. This creates significant ambiguity for agent selection.

Naming Consistency2/5

The naming conventions are inconsistent and chaotic. There's a mix of snake_case ('brave_search_summary') and lowercase with underscores ('search_brave_with_summary'), plus inconsistent verb usage ('get_' vs just the noun 'url_content'). The bilingual naming (English and Chinese) further compounds the inconsistency with no discernible pattern.

Tool Count3/5

With 6 tools, the count is reasonable for a search/content retrieval server. However, since 4 of these tools are essentially duplicates (2 search pairs and 2 URL content pairs), the effective tool count is much lower, making the server feel thin despite the nominal count being appropriate.

Completeness3/5

The server covers basic search and content retrieval operations, but there are notable gaps. While it provides search (general and news) and URL fetching, there's no way to filter search results, handle pagination, or access more advanced Brave Search features. The duplication of tools doesn't add functional coverage, leaving the surface incomplete for robust web interaction.

Maintenance

ActivityInactive
ResponsivenessNo issues

Resources