Filesystem MCP
Provides secure filesystem access for AI assistants with optimizations like file reading limits and depth-limited traversal to improve token efficiency. It enables AI models to read, write, and search files within explicitly allowed directories while automatically skipping large system folders.
README
Filesystem MCP
A Model Context Protocol (MCP) server that provides secure filesystem access to AI assistants in Code Maestro Desktop.
What it does
This MCP server allows AI assistants to:
- Read and write files
- Create and navigate directories
- Search for files
- Edit existing files with precision
- And more filesystem operations
All operations are restricted to the directories you explicitly allow.
Why This Fork?
We forked the original @modelcontextprotocol/server-filesystem to make it more efficient for AI assistants. The original package could return extremely large outputs that waste tokens and overwhelm language models.
Our improvements:
- Smart file reading limits - Prevents reading gigantic files in full
- Directory listing caps - Returns manageable results even in folders with thousands of files
- Depth-limited tree traversal - Avoids getting lost in deep folder structures
- Optimized search results - Returns relevant matches without flooding the output
- Skip large folders - Automatically bypasses
node_modules,.git, and similar directories
These changes ensure your AI assistant gets the information it needs without drowning in unnecessary data.
Quick Start
1. Open CM Desktop and click connectors
2. Click "Add" and paste the following JSON:
{
"mcpServers": {
"Code Maestro - Filesystem MCP": {
"transport": "stdio",
"enabled": true,
"command": "npx",
"args": [
"github:codemaestroai/filesystem-mcp",
"<YOUR_PROJECT_PATH>"
],
"env": {},
"url": null,
"headers": null
}
}
}
3. Replace <YOUR_PROJECT_PATH> with the actual path to your project directory
Example paths:
- Windows:
"C:\\Users\\YourName\\Projects\\my-project" - macOS/Linux:
"/home/username/projects/my-project"
You can specify multiple directories:
"args": [
"github:codemaestroai/filesystem-mcp",
"/path/to/project1",
"/path/to/project2"
]
Security
- The MCP server can ONLY access directories you explicitly specify
- All file operations are logged
- Attempts to access files outside allowed directories will be rejected
Note: This is a temporary solution. We're working on native file operation support in CM Desktop.
Recommended Servers
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.
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.
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.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
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.
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.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.
Neon Database
MCP server for interacting with Neon Management API and databases
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.
E2B
Using MCP to run code via e2b.