Local File Reader

🚀 MCP Local File Reader

A local file reading server based on the Model Context Protocol (MCP), allowing AI models to securely access the local file system.

🚀 Quick Start

The MCP Local File Reader is a server that enables AI models to interact with the local file system. It provides a set of tools for listing and reading files securely.

✨ Features

Resources

• List and access files in the local file system.
• Support multiple file types, including text and binary files.
• Automatically detect the MIME type of files and provide appropriate content handling.

Tools

• read_file - Read the content of a specified file.
  • Return the full content for text files.
  • Return a file information digest for binary files.
• list_files - List all files in a specified directory.
  • Return a list of file names.
• get_file_info - Get detailed information about a specified file.
  • Return metadata such as file size, type, and creation time.

📦 Installation

npm install mcp-local-file-reader

💻 Usage Examples

As a Command-Line Tool

After installation, you can run it directly in the command line:

npx mcp-local-file-reader

This will start an MCP node server, listening for LLM conversations.

Integration with AI Tools

Using in Windsurf

In Windsurf, you don't need to manually start mcp-local-file-reader. Just configure it, and Windsurf will start mcp-local-file-reader automatically (configure in mcp_config.json of Windsurf and click refresh to update the server).

1. Add the following configuration to the mcp_config.json configuration file in Windsurf (temporarily no need for FILE_TOOLS_API_KEY):

Local Deployment

{
  "servers": {
    "file-tools": {
      "command": "cmd",
      "args": [
        "/c",
        "node",
        "path/to/mcp-local-file-reader/build/index.js"
      ],
      "env": {
        "FILE_TOOLS_API_KEY": ""
      }
    }
  }
}

Using npm Deployment

{
  "servers": {
    "file-tools": {
      "command": "cmd",
      "args": [
        "/c",
        "npx",
        "mcp-local-file-reader"
      ],
      "env": {
        "FILE_TOOLS_API_KEY": ""
      }
    }
  }
}

Click the hammer icon in the Windsurf dialog and then click refresh to update the server. Official configuration instructions: https://docs.codeium.com/windsurf/mcp

2. Restart Windsurf to load the new MCP configuration.

Using in Other MCP-Compatible AI Applications

Follow the MCP integration guide of the specific application to add this server as a tool provider. The configuration in Cursor is similar to that in Windsurf, and it may be a bit more complex in VSCode. But essentially, it's all about configuring mcp.json, and you can refer to similar configurations.

🔧 Technical Details

Development

Install dependencies:

npm install

Build the server:

npm run build

Development mode (auto-rebuild):

npm run watch

Test with MCP Inspector:

npm run inspector

⚠️ Important Note

This server allows AI models to access the local file system. Please use it with caution and ensure:

1. Only allow access to necessary directories.
2. Do not expose sensitive files or directories.
3. Consider adding additional security restrictions before using it in a production environment.

📄 License

MIT