RemoteZip MCP Server
Enables accessing and extracting files from remote zip archives over HTTP, HTTPS, and FTP without downloading the entire archive. Supports listing files, extracting individual files, and getting file information using partial reading techniques.
README
RemoteZip MCP Server
An MCP server that provides tools for accessing remote zip files over HTTP, HTTPS, and FTP protocols without downloading the entire archive.
Features
- List files in remote zip archives
- Extract individual files from remote zip archives
- Get file information and statistics
- Support for HTTP, HTTPS, and FTP protocols
- Partial reading - only downloads necessary parts of the zip file
Installation
From Source
- Install Python 3.10 or higher
- Clone the repository:
git clone https://github.com/yourusername/remotezip-mcp.git cd remotezip-mcp - Install dependencies:
pip install -e .
From GitHub Releases
Download the latest release from the Releases page and install:
pip install remotezip_mcp-1.0.0-py3-none-any.whl
MCP Server Configuration
For VSCode
-
Install the MCP extension:
- Search for "MCP (Model Context Protocol)" in VSCode extensions
- Install the extension by Anthropic
-
Create MCP configuration file:
- Create the file
~/.vscode/mcp.json(on macOS/Linux) or%APPDATA%\Code\User\mcp.json(on Windows) - Or use the global config:
~/.config/mcp/mcp.json
- Create the file
-
Add the following configuration to your mcp.json:
{
"mcpServers": {
"remotezip": {
"command": "remotezip-mcp",
"args": [],
"env": {
"PATH": "$PATH"
}
}
}
}
- Restart VSCode to load the MCP server
For Cursor
-
Open Cursor settings:
- Press
Cmd/Ctrl + ,to open settings - Or go to File → Preferences → Settings
- Press
-
Navigate to MCP settings:
- Search for "mcp" in the settings search bar
- Look for "MCP: Servers" or "Model Context Protocol" section
-
Add a new MCP server with these exact settings:
- Name:
RemoteZip MCP Server - Command:
remotezip-mcp - Arguments: Leave this field empty
- Environment Variables: (optional)
- Save and restart Cursor to activate the MCP server
Alternative: Manual MCP Configuration
If your editor doesn't have built-in MCP support, you can also run the server manually:
# Start the MCP server
remotezip-mcp
Then connect to it using your MCP client with the server running on the default port.
Troubleshooting
- Command not found: Make sure
remotezip-mcpis in your PATH after installation - Permission denied: Try running with
sudoor check file permissions - Server won't start: Verify the installation with
remotezip-mcp --help - Configuration not loading: Check the JSON syntax in your mcp.json file
Usage
Run the server:
python remotezip_server.py
Or use the installed command:
remotezip-mcp
Tools
list_files(url): Get list of files in the archiveextract_file(url, filename, local_path): Extract a file to local storageget_file_info(url, filename): Get details about a specific fileget_zip_statistics(url): Get overall statistics of the archive
Development
Setup Development Environment
# Create virtual environment
python -m venv .venv
source .venv/bin/activate # On Windows: .venv\Scripts\activate
# Install in development mode
pip install -e .
Running Tests
# Run the MCP server
python remotezip_server.py
Creating Releases
Use the provided release script:
./release.sh 1.0.0
This will:
- Create a git tag
r1.0.0 - Push the tag to GitHub
- Trigger GitHub Actions to build and release
GitHub Actions
The repository includes automated release workflows:
- Release Workflow (
.github/workflows/release.yml): Automatically builds and releases when version tags are pushed - Triggers on tags matching pattern
r* - Builds Python wheel and source distribution
- Creates GitHub release with generated release notes
Dependencies
remotezip: For partial reading of remote zip filesmcp: Model Context Protocol frameworkfastmcp: FastMCP server implementation
License
MIT License - see LICENSE file for details
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。