mark3labs/mcp-filesystem-server

Overview

A Go-based server that implements the Model Context Protocol (MCP) to expose filesystem operations as tools for AI agents. It allows AI models to read, write, list, and modify files and directories through a standardized interface.

Best for

Best for
Developers building AI agents that need direct filesystem interaction

Use cases

• Enabling LLM agents to read and write local files
• Providing file search and directory listing for AI workflows
• Integrating file management into MCP-compatible assistant tools

How to use

Install

go install github.com/mark3labs/mcp-filesystem-server@latest

Tools exposed

• read_file
• read_multiple_files
• write_file
• copy_file
• move_file
• delete_file
• modify_file
• list_directory
• create_directory
• tree
• search_files
• search_within_files
• get_file_info
• list_allowed_directories

Example client config

{\n  "mcpServers": {\n    "filesystem": {\n      "command": "mcp-filesystem-server",\n      "args": ["/path/to/allowed/directory", "/another/allowed/directory"]\n    }\n  }\n}

Notes

A Go-based server that implements the Model Context Protocol (MCP) to expose filesystem operations as tools for AI agents. It allows AI models to read, write, list, and modify files and directories through a standardized interface.

646 stars on GitHub. Last updated 2025-11-24. Licensed MIT.

Use cases

• Enabling LLM agents to read and write local files
• Providing file search and directory listing for AI workflows
• Integrating file management into MCP-compatible assistant tools

Pros

• Lightweight Go server with minimal dependencies
• Standard MCP interface ensures broad compatibility
• Supports essential filesystem operations out of the box

Cons

• Requires running a separate server process on the host
• No built-in authentication or access control
• Limited to local filesystem; no remote or cloud storage support

Indexed from awesome-mcp-servers-punkpeye and enriched against its public facts.