FS Context MCP Server

FS Context MCP Server

A secure, read-only MCP server that enables filesystem scanning, searching with glob patterns and regex, file reading, and batch operations with comprehensive path validation and security protections.

Category
Visit Server

README

Filesystem MCP

npm version License Node.js Version Docker Image

Install in VS Code Install in VS Code Insiders Install in Claude Desktop Install in Cursor

MCP Server that enables LLMs to interact with the local filesystem. Provides tools for navigation, file management, searching, and analysis, all within a secure, allowed set of directories. Ideal for agents needing to read/write files, explore directory structures, or perform file operations as part of their tasks.

Key Features

  • Navigation: List directories (ls), view trees (tree), and discover workspace roots (roots).
  • File Management: Create (mkdir), write (write), edit (edit), move (mv), and delete (rm) files/directories.
  • Search: Find files by glob pattern (find) or search content using regex (grep).
  • Analysis: Inspect metadata (stat), calculate hashes (calculate_hash), and compare files (diff_files).
  • Batch Operations: Read multiple files (read_many) or replace text across many files (search_and_replace).
  • Security: Strictly scoped to allowed directories provided at startup.

Tech Stack

  • Runtime: Node.js >= 24
  • Language: TypeScript
  • SDK: @modelcontextprotocol/sdk
  • Libraries: zod, commander, re2, diff

Repository Structure

.
├── src/
│   ├── index.ts        # Entry point
│   ├── server.ts       # MCP Server implementation
│   ├── tools/          # Individual tool implementations
│   └── lib/            # Shared utilities
├── assets/             # Images and static resources
├── scripts/            # Build and maintenance scripts
└── package.json

Requirements

  • Node.js >= 24

Quickstart

Run directly with npx:

npx -y @j0hanz/filesystem-mcp@latest "C:\path\to\allowed\directory"

Installation

NPX (Recommended)

npx -y @j0hanz/filesystem-mcp@latest [options] [directories...]

Docker

docker run -i --rm \
  -v /path/to/your/project:/projects/workspace:ro \
  ghcr.io/j0hanz/filesystem-mcp:latest \
  /projects/workspace

Mount host directories as volumes to /projects/ and pass the container paths as arguments.

From Source

  1. Clone the repository

  2. Install dependencies:

    npm ci
    
  3. Build the project:

    npm run build
    
  4. Run:

    node dist/index.js [options] [directories...]
    

Configuration

The server requires specifying allowed directories via command-line arguments.

Arguments

Argument Description
[allowedDirs...] Positional arguments specifying the root directories the server can access.

Options

Option Description
--allow-cwd Allow the current working directory as an additional root.
-v, --version Display server version.
-h, --help Display command help.

Usage

Stdio Transport

The server communicates via stdio. Ensure your MCP client is configured to run the server command and capture standard input/output.

MCP Surface

Tools

Tool Description Key Parameters
roots List allowed workspace roots None
ls List directory contents path, includeHidden
find Find files by glob pattern pattern, path, maxDepth
tree Generate directory tree path, maxDepth
read Read file content path, head
read_many Read multiple files paths
grep Search file content (regex/literal) pattern, path, isRegex
stat Get file metadata path
stat_many Get metadata for multiple files paths
calculate_hash Calculate SHA-256 hash path
mkdir Create directory (recursive) path
write Write file (create/overwrite) path, content
edit Edit file (string replacement) path, edits
mv Move or rename file/directory source, destination
rm Delete file or directory path, recursive
diff_files Generate unified diff original, modified
apply_patch Apply unified patch path, patch
search_and_replace Search & replace across files filePattern, searchPattern, replacement

Resources

URI Pattern Description
internal://instructions Usage guidance and documentation
filesystem-mcp://result/{id} Ephemeral cached tool output (for large results)

Prompts

Prompt Description
get-help Returns usage instructions for the server

Client Configuration Examples

<details> <summary><strong>Claude Desktop</strong></summary>

Add to your claude_desktop_config.json:

{
  "mcpServers": {
    "filesystem": {
      "command": "npx",
      "args": [
        "-y",
        "@j0hanz/filesystem-mcp@latest",
        "C:\\path\\to\\allowed\\directory"
      ]
    }
  }
}

</details>

<details> <summary><strong>Cursor</strong></summary>

Add to your .cursor/mcp.json or configure via UI:

{
  "id": "filesystem",
  "name": "filesystem",
  "command": "npx",
  "args": ["-y", "@j0hanz/filesystem-mcp@latest", "${workspaceFolder}"]
}

</details>

<details> <summary><strong>VS Code</strong></summary>

Add to your .vscode/mcp.json:

{
  "mcpServers": {
    "filesystem": {
      "command": "npx",
      "args": ["-y", "@j0hanz/filesystem-mcp@latest", "${workspaceFolder}"]
    }
  }
}

</details>

<details> <summary><strong>Docker (any MCP client)</strong></summary>

Use the Docker image with any MCP client that supports stdio transport:

{
  "mcpServers": {
    "filesystem": {
      "command": "docker",
      "args": [
        "run", "-i", "--rm",
        "-v", "/path/to/project:/projects/workspace:ro",
        "ghcr.io/j0hanz/filesystem-mcp:latest",
        "/projects/workspace"
      ]
    }
  }
}

</details>

<details> <summary><strong>Codex</strong></summary>

Add to your configuration:

[mcp_servers.filesystem-mcp]
command = "npx"
args = ["-y", "@j0hanz/filesystem-mcp@latest", "${workspaceFolder}"]

</details>

Security

  • Path Restrictions: All file operations are strictly validated against the allowed root directories provided at startup.
  • Path Validation: Uses isPathWithinDirectories to prevent path traversal attacks.
  • Hidden Files: Hidden files (starting with .) are excluded by default in listings and searches unless explicitly requested.

Development Workflow

  1. Install Dependencies:

    npm install
    
  2. Development Mode (watch):

    npm run dev
    
  3. Run Locally:

    npm start -- --allow-cwd
    
  4. Test:

    npm run test
    
  5. Lint & Format:

    npm run lint
    npm run format
    

License

MIT

Recommended Servers

playwright-mcp

playwright-mcp

A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.

Official
Featured
TypeScript
Magic Component Platform (MCP)

Magic Component Platform (MCP)

An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.

Official
Featured
Local
TypeScript
Audiense Insights MCP Server

Audiense Insights MCP Server

Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.

Official
Featured
Local
TypeScript
VeyraX MCP

VeyraX MCP

Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.

Official
Featured
Local
Kagi MCP Server

Kagi MCP Server

An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.

Official
Featured
Python
graphlit-mcp-server

graphlit-mcp-server

The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.

Official
Featured
TypeScript
E2B

E2B

Using MCP to run code via e2b.

Official
Featured
Neon Database

Neon Database

MCP server for interacting with Neon Management API and databases

Official
Featured
Exa Search

Exa Search

A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.

Official
Featured
Qdrant Server

Qdrant Server

This repository is an example of how to create a MCP server for Qdrant, a vector search engine.

Official
Featured