LocalFiles MCP Server
Provides secure access to local files from specified directories. Enables listing files, reading contents, and configurable file size and extension filtering.
README
LocalFiles MCP Server
A Model Context Protocol (MCP) server that provides secure access to local files from specified directories.
Features
- List files in configured directories
- Read file contents (text and binary)
- File size and extension filtering
- Secure path validation
- Configuration via .env file or environment variables
Configuration
Method 1: .env file (Recommended)
Create a .env file in the same directory as app.py:
# Comma-separated list of allowed directories
ALLOWED_DIRECTORIES=/home/user/documents,/home/user/projects,/path/to/your/files
# Maximum file size in bytes (default: 10MB)
MAX_FILE_SIZE=10485760
# Allowed file extensions (comma-separated)
ALLOWED_EXTENSIONS=.txt,.md,.py,.js,.json,.yaml,.yml,.csv,.xml,.html,.css
Method 2: Environment Variables
Set environment variables directly:
export ALLOWED_DIRECTORIES="/home/user/documents,/home/user/projects"
export MAX_FILE_SIZE=10485760
export ALLOWED_EXTENSIONS=".txt,.md,.py,.js,.json,.yaml,.yml,.csv,.xml,.html,.css"
Usage
Start the server
python app.py
Available Tools
- lf_list_files - List files in allowed directories
- lf_read_file - Read contents of a specific file
- lf_get_config - Get current server configuration
Security
- Only files within configured
ALLOWED_DIRECTORIEScan be accessed - File size limits prevent reading overly large files
- File extension filtering restricts access to specified file types
- All paths are resolved and validated to prevent directory traversal attacks
Troubleshooting
"No allowed directories configured" error
This error occurs when:
- No .env file exists - Create a
.envfile withALLOWED_DIRECTORIESsetting - Empty ALLOWED_DIRECTORIES - Ensure the value is not empty
- Wrong working directory - The server looks for
.envin the script directory first, then current directory - Environment variables not set - If not using .env, ensure environment variables are properly set
Debug configuration
Use the lf_get_config tool to check what configuration the server is actually using:
# This will show the current configuration
lf_get_config()
Example .env file
# Example configuration for a development environment
ALLOWED_DIRECTORIES=/home/alice/projects,/home/alice/documents,/tmp/safe-files
MAX_FILE_SIZE=52428800
ALLOWED_EXTENSIONS=.txt,.md,.py,.js,.json,.yaml,.yml,.csv,.xml,.html,.css,.log
Requirements
- Python 3.7+
- fastmcp library
- pathlib (built-in)
Installation
pip install fastmcp
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。