AI File Assistant MCP
MCP server providing AI-powered file operations including read, write, edit, search, and file management via Model Context Protocol. Also includes a Flask web app with file manager and user admin.
README
AI File Assistant MCP Server
An MCP (Model Context Protocol) server that provides AI-powered file operations: read, write, search, edit, and manage files.
Also includes a Flask web app with a full file manager, user management, and admin panel.
Features
MCP Server (server.py)
- Read files with line numbers
- Write files with automatic directory creation
- Edit files via exact string replacement
- Search file contents using regex
- Glob file patterns for discovery
- List directories
- Move files
- Copy files
- Delete files
- Get file metadata
Web App (app.py)
- File Manager - Browse, upload, download, preview, search, and delete files via browser
- File Preview - View text, images, audio, and video files inline
- Search - Search file names and content directly from the web UI
- User Authentication - Register, login, profile management
- Admin Panel - Manage users (activate/deactivate, promote, delete)
- Activity Dashboard - Charts showing daily activity and action breakdown
- File Stats - Track file count and storage usage per user
- Rate Limiting - Login brute force protection
- Drag & Drop Upload - Upload multiple files with drag and drop
Installation
pip install -r requirements.txt
Configuration
Copy .env (already provided) and customize:
| Variable | Default | Description |
|---|---|---|
SECRET_KEY |
(auto-generated) | Flask session secret key |
ADMIN_PASSWORD |
admin123 |
Default admin password |
ADMIN_EMAIL |
admin@fileassistant.local |
Default admin email |
FLASK_PORT |
5000 |
Web server port |
UPLOAD_FOLDER |
./uploads |
File storage directory |
MAX_UPLOAD_SIZE |
50 |
Max upload size in MB |
Usage
MCP Server (stdio)
python server.py
For development mode with MCP Inspector:
python server.py dev
Web App
python app.py
Open http://localhost:5000 in your browser.
MCP Client Example
python client.py
Testing
python tests.py
Default Admin Credentials
- Username: admin
- Password: admin123
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。