Skip to main content
Glama
VitexSoftware

mcp-server-filesystem

mcp-server-filesystem

MCP 服务器,将 Linux 文件系统操作作为工具暴露给 LLM 代理,基于 FastMCP 构建。设计为 mcprack MCP 代理/目录工具的轻量级、低依赖测试服务器。

配置

无需凭据。所有配置均通过环境变量完成,在启动时读取一次:

变量

必需

描述

FS_ROOT

所有路径被限制在其中的目录(默认:启动时的当前工作目录)

FS_READONLY

true(默认)阻止修改类工具;设置为 false 以允许写入/删除/移动

所有提供给工具的路由都相对于 FS_ROOT 解析,如果会逃逸出该目录(通过 .. 或符号链接)则被拒绝。这是应用级边界,而非加固的沙箱(无 chroot/挂载命名空间)。

Related MCP server: ab_mcp

工具

只读:list_dirread_filestatexistsglob_search

修改类(需要 FS_READONLY=false):write_filemkdirdeletemovecopy

用法

推荐 — 来自 VitexSoftware 仓库的 Debian 包

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

backports 组件是必需的,而非可选的:python3-mcppython3-fastmcp 的依赖)需要 python3-jsonschema >= 4.20.0,这比 Debian trixie 在 main 中提供的版本更新——它仅在 backports 中可用。没有它,apt install 会因 python3-jsonschema 的依赖未满足而失败。

这将一步安装 python3-fastmcpmcp-server-filesystem

备选 — pip

pip install -e .
export FS_ROOT=/path/to/sandbox
mcp-server-filesystem

mcprack 集成

mcprack 中注册此服务器:

  • 命令:mcp-server-filesystem

  • 环境变量:FS_ROOT=/path/to/sandboxFS_READONLY=false(如果需要写入)

mcprack 将其作为 stdio 子进程启动,并按用户通过 HTTP 代理。

开发

pytest
A
license - permissive license
Not graded
quality - not tested
C
maintenance

Maintenance

Maintainers
Response time
Release cycle
Releases (12mo)
Commit activity

Resources

Unclaimed servers have limited discoverability.

Looking for Admin?

If you are the server author, to access and configure the admin panel.

Related MCP Servers

  • -
    license
    Not graded
    quality
    Not graded
    maintenance
    A Filesystem MCP server that allows an LLM to read and list files from a specified directory on your local machine through the Model Context Protocol.
    2
  • A
    license
    A
    quality
    C
    maintenance
    A file system MCP server that confines AI assistant file tools to a single folder, with read-only by default and optional write access.
    7
    MIT
  • A
    license
    A
    quality
    B
    maintenance
    Provides LLMs with local filesystem operations (read/write files, list directories) and command execution via MCP, enabling file management and task automation within AI clients.
    7
    17
    ISC

View all related MCP servers

Related MCP Connectors

  • Agent-native MCP server over the public saagarpatel.dev corpus. Read-only, stateless.

  • OCR, transcription, file extraction, and image generation for AI agents via MCP.

  • MCP server for secureFlows: token-free URL builders and integration-linting tools for AI agents.

View all MCP Connectors

Latest Blog Posts

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/VitexSoftware/mcp-server-filesystem'

If you have feedback or need assistance with the MCP directory API, please join our Discord server