mcp-server-filesystem
Exposes Linux filesystem operations as tools for LLM agents, with configurable root directory and read-only mode, built with FastMCP.
README
<img src="debian/mcp-server-filesystem.svg" width="96" height="96" alt="mcp-server-filesystem icon">
mcp-server-filesystem
MCP server exposing Linux filesystem operations as tools for LLM agents, built with FastMCP. Designed as a lightweight, low-dependency test server for the mcprack MCP proxy/catalog tool.
Configuration
No credentials are needed. All configuration is via environment variables, read once at startup:
| Variable | Required | Description |
|---|---|---|
FS_ROOT |
no | Directory all paths are confined to (default: current working directory at startup) |
FS_READONLY |
no | true (default) blocks mutating tools; set to false to allow writes/deletes/moves |
All paths given to tools are resolved against FS_ROOT and rejected if they
would escape it (via .. or a symlink). This is an application-level
boundary, not a hardened sandbox (no chroot/mount namespace).
Tools
Read-only: list_dir, read_file, stat, exists, glob_search.
Mutating (require FS_READONLY=false): write_file, mkdir, delete,
move, copy.
Usage
Recommended — Debian package from VitexSoftware repository
sudo curl -fsSL http://repo.vitexsoftware.com/KEY.gpg -o /usr/share/keyrings/vitexsoftware-archive-keyring.gpg
echo "deb [signed-by=/usr/share/keyrings/vitexsoftware-archive-keyring.gpg] http://repo.vitexsoftware.com trixie main backports" \
| sudo tee /etc/apt/sources.list.d/vitexsoftware.list
sudo apt update
sudo apt install mcp-server-filesystem
The backports component is required, not optional: python3-mcp (a
python3-fastmcp dependency) needs python3-jsonschema >= 4.20.0, which is
newer than the version Debian trixie ships in main — it's only available
in backports. Without it, apt install fails with an unmet-dependency
error on python3-jsonschema.
This installs python3-fastmcp and mcp-server-filesystem in a single step.
Alternative — pip
pip install -e .
export FS_ROOT=/path/to/sandbox
mcp-server-filesystem
mcprack integration
Register this server in mcprack with:
- Command:
mcp-server-filesystem - Env:
FS_ROOT=/path/to/sandbox,FS_READONLY=false(if writes are needed)
mcprack spawns it as a stdio subprocess and proxies it over HTTP per user.
Development
pytest
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。