Skip to main content
Glama

Filesystem MCP Server

Filesystem MCP Server (v0.1)

A simple MCP server for filesystem operations with directory sandboxing and read-only mode support.

Early version (0.1) - not thoroughly tested/reviewed, use with caution

Quick Start

Install dependencies:

git clone <repo_url> uv sync

Run the server:

# Normal mode - full filesystem access uv run main.py /path/to/allowed/directory # Read-only mode - no write operations uv run main.py --read-only /path/to/allowed/directory

Related MCP server: MCP File System Server

Features

  • Directory sandboxing (operations restricted to allowed paths)

  • --read-only flag disables write operations

  • Gitignore support for searches

  • Human-readable error messages

Usage Examples

Normal Mode

uv run main.py ~/projects # Provides: create_file, delete_file, move_file, edit_file, create_directory + read tools

Read-Only Mode

uv run main.py --read-only ~/projects # Provides only: read_text_file, list_directory, search_files, grep, directory_tree

Multiple Directories

uv run main.py /home/user/docs /tmp/workspace uv run main.py --read-only /var/log /etc/config

Available Tools

Read Operations (always available):

  • read_text_file - Read file contents

  • list_directory - List directory contents

  • directory_tree - JSON directory structure

  • search_files - Find files by pattern

  • grep - Search text in files

  • read_multiple_files - Bulk file reading

Write Operations (disabled in

  • create_file - Create new files

  • delete_file - Delete files

  • move_file - Move/rename files

  • edit_file - Edit file contents

  • create_directory - Create directories

Dependencies

From pyproject.toml:

  • fastmcp>=2.11.2 - MCP framework

  • pathspec>=0.12.1 - Gitignore patterns

Security Notes

  • All paths resolved with Path.resolve() to prevent traversal

  • Operations restricted to allowed directories only

  • Binary files rejected for text operations

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/Preston-Harrison/fs-mcp-py'

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