WebSearch MCP Server
一个基于 MCP (Model Context Protocol) 的智能网页搜索服务器,支持 AI 增强搜索、Brave 搜索和网页抓取功能。
推荐: 本项目推荐使用
SOTASearch.py,它结合了 AI 深度搜索和 Brave 搜索,并行执行两种搜索以提升速度。适合与 CherryStudio 配合使用。
功能
| 工具 | 说明 |
|---|---|
web_search |
AI 深度搜索 + Brave 搜索(并行执行),返回链接列表和 AI 总结 |
fetch_html |
抓取网页 HTML 内容 |
fetch_text |
抓取网页并提取纯文本 |
fetch_metadata |
抓取网页元数据(标题、描述、链接) |
安装
1. 克隆项目
git clone https://github.com/yourusername/WebSearchMCP.git
cd WebSearchMCP
2. 安装依赖
使用 uv(推荐):
uv sync
使用 pip:
pip install -r requirements.txt
3. 配置环境变量
复制 .env.example 为 .env 并填入配置:
cp .env.example .env
主要配置项:
# 代理配置(可选)
PROXY=http://127.0.0.1:7890
CF_WORKER=https://your-worker.workers.dev
# OpenAI API 配置(SOTASearch 必需)
OPENAI_API_KEY=sk-xxx
OPENAI_BASE_URL=https://api.openai.com/v1
OPENAI_MODEL=gpt-4o
快速开始
# 使用 uv(推荐)
uv run SOTASearch.py
# 或使用 python
python SOTASearch.py
配置 CherryStudio
在 CherryStudio 的 MCP 服务器设置中添加(必须使用虚拟环境中的 Python 绝对路径):
Windows 示例:
{
"mcpServers": {
"sota-search": {
"name": "Websearch",
"type": "stdio",
"command": "D:\\Code\\github\\WebSearchMCP\\.venv\\Scripts\\python.exe",
"args": ["D:/Code/github/WebSearchMCP/SOTASearch.py"]
}
}
}
macOS/Linux 示例:
{
"mcpServers": {
"sota-search": {
"name": "Websearch",
"type": "stdio",
"command": "/path/to/WebSearchMCP/.venv/bin/python",
"args": ["/path/to/WebSearchMCP/SOTASearch.py"]
}
}
}
注意: 推荐使用项目虚拟环境中的 Python 解释器(
.venv/Scripts/python.exe或.venv/bin/python),确保依赖正确加载。系统 Python 也可以使用,但需确保已安装所有依赖。
提示: 代理和 API 配置建议写在
.env文件中,无需在命令行参数中指定。
命令行参数
| 参数 | 说明 | 示例 |
|---|---|---|
--proxy |
本地代理地址 | --proxy http://127.0.0.1:7890 |
--cf-worker |
Cloudflare Worker 地址 | --cf-worker https://xxx.workers.dev |
--openai-api-key |
OpenAI API Key | --openai-api-key sk-xxx |
--openai-base-url |
OpenAI API Base URL | --openai-base-url https://api.openai.com/v1 |
--openai-model |
模型名称(默认 gpt-4o) | --openai-model gpt-4o |
工具说明
web_search
AI 深度搜索 + Brave 搜索(并行执行)。
参数:
query: 搜索关键词(必填)max_results: 返回的最大结果数,默认 10
返回格式:
{
"success": true,
"query": "搜索关键词",
"links": [
{"title": "标题", "url": "链接", "description": "描述"}
],
"ai_summary": "AI 总结内容"
}
fetch_html
抓取网页的原始 HTML 内容。
参数:
url: 目标网址(必填)headers: 可选的请求头
fetch_text
抓取网页并提取纯文本内容(去除 HTML 标签)。
参数:
url: 目标网址(必填)headers: 可选的请求头
fetch_metadata
抓取网页的元数据信息。
参数:
url: 目标网址(必填)headers: 可选的请求头
返回:标题、描述、页面链接列表
特性
- AI 深度搜索 + Brave 搜索并行执行
- 支持本地代理 (
--proxy) - 支持 Cloudflare Worker 代理 (
--cf-worker) - 自动内容截断,防止响应过大
- 支持
.env文件配置
依赖
- Python 3.10+
- fastmcp
- beautifulsoup4
- curl_cffi
- lxml
- openai
Metadata
Release files for iflow-mcp_vonequinox-websearchmcp 0.1.0
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | Provenance | Actions |
|---|---|---|---|---|
| iflow_mcp_vonequinox_websearchmcp-0.1.0.tar.gz | 13.8 kB | Without provenance | Details for iflow_mcp_vonequinox_websearchmcp-0.1.0.tar.gz |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Actions Reset |
|---|---|---|---|---|
| iflow_mcp_vonequinox_websearchmcp-0.1.0-py3-none-any.whl | Python 3 | none | any | Details for iflow_mcp_vonequinox_websearchmcp-0.1.0-py3-none-any.whl |
Total release size: 26.8 kB
Release files / iflow_mcp_vonequinox_websearchmcp-0.1.0.tar.gz
| Download URL | iflow_mcp_vonequinox_websearchmcp-0.1.0.tar.gz |
|---|---|
| Size | 13.8 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
d0d4a36ac95420201d2ba9345318d496e4da1e5b63b519fbd526d6db652444dc
|
|
BLAKE2b-256 checksum How to use checksums |
fa7c5ce44a0c2bbf4eadf8b50079f5e3ffa6616bbc303a4360b2243b07dd2ee6
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
uv/0.10.0 {"installer":{"name":"uv","version":"0.10.0","subcommand":["publish"]},"python":null,"implementation":{"name":null,"version":null},"distro":{"name":"Debian GNU/Linux","version":"13","id":"trixie","libc":null},"system":{"name":null,"release":null},"cpu":null,"openssl_version":null,"setuptools_version":null,"rustc_version":null,"ci":null}
|
Release files / iflow_mcp_vonequinox_websearchmcp-0.1.0-py3-none-any.whl
| Download URL | iflow_mcp_vonequinox_websearchmcp-0.1.0-py3-none-any.whl |
|---|---|
| Size | 12.9 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
020bffa97fa8e624eda770a04370ffee7ab4cf25c555581f7c6cc78c86e24961
|
|
BLAKE2b-256 checksum How to use checksums |
ec9c38e4f6557c0aca66fa76860fb3ff10ae3c1f201d5608c3ed57b3a50e4dda
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
uv/0.10.0 {"installer":{"name":"uv","version":"0.10.0","subcommand":["publish"]},"python":null,"implementation":{"name":null,"version":null},"distro":{"name":"Debian GNU/Linux","version":"13","id":"trixie","libc":null},"system":{"name":null,"release":null},"cpu":null,"openssl_version":null,"setuptools_version":null,"rustc_version":null,"ci":null}
|
Security for iflow-mcp_vonequinox-websearchmcp 0.1.0
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
No provenance found
None of the files in this release were uploaded with provenance metadata.
A note on package security
PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.
To learn more about protecting yourself from malicious actors, visit our package security documentation.