FileSystem

Local MCP server for Visual Studio 2022 that provides code-workspace access by granting AI agents selective access to project folders and files.

python

3

GitHub Stars

python

Language

6 months ago

First Indexed

2 months ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

{
  "mcpServers": {
    "oncorporation-filesystem_server": {
      "command": "python",
      "args": [
        "/absolute/path/to/your/project/filesystem_server/app.py"
      ]
    }
  }
}

You run a local MCP Server that integrates with Visual Studio 2022 to give AI agents selective access to your project folders and files. It supports debugging in VS, safe directory/file access, and cross‑platform path handling so you can develop locally without complex workspace setups.

How to use

Use the server with an MCP client to perform filesystem operations such as listing directory contents, reading files, and validating access to configured directories. You can run the server locally in stdio mode (a local process) or connect to it via an MCP client using command-line arguments. The server prioritizes command-line arguments when provided, and falls back to configuration files for debugging in Visual Studio 2022.

How to install

Prerequisites: Python 3.10 or later installed on your machine.

Clone or download the project directory that contains the filesystem server.

Install dependencies using the UV package manager.

uv sync

Configuration and usage notes

Choose one of the following configuration approaches. You can combine them using the Hybrid Approach, where command-line arguments take precedence over config.json when both are present.

Configuration options (summary)

Option 1: Config.json File (Simplest, recommended for beginners) Create a config.json with allowed directories and extensions. For debugging in VS 2022, place config.json next to app.py. For MCP server usage, place config.json next to the .mcp.json file.

Option 2: Command-Line Arguments (Advanced)

Configure the server entirely via command-line arguments. This makes the MCP client configuration self-contained and easier to version-control.

Option 3: Hybrid Approach

The server uses command-line arguments when provided, then falls back to config.json. This gives you the best of both worlds: efficiency for MCP clients and ease of debugging in Visual Studio.

Troubleshooting

If you encounter issues, validate your directory paths and permissions first. Use the init tool to verify accessibility of configured directories, then retry the operation that failed.

Security

Access is restricted to explicitly configured directories and file extensions. All paths are validated before access, and the server runs with your user permissions, making it suitable for secure local development.

Cross-Platform Path Support

The server normalizes Windows and Unix style paths automatically so you can mix separators without errors. It accepts both backslashes and slashes in paths.

Notes for Visual Studio 2022 debugging

For clean debugging, ensure a config.json is present next to app.py. Press F5 to start debugging, and use breakpoints as you would with any Python application.

Advanced & real-world configuration examples

Refer to the provided examples for how to structure MCP configurations when running from Visual Studio 2022 or via an MCP client. Use the same command patterns shown in the examples to ensure proper startup behavior.

Available tools

init

Validates accessibility of configured directories and can optionally list a directory and/or read a file.

list_directory

Lists files and subdirectories in a given directory with optional progress reporting.

read_file

Reads the content of a specified text file.

read_file_binary

Reads a file as base64-encoded binary content.

list_resources

Lists all resources (files and directories) in a directory or across all allowed directories in MCP resource format.

get_resource

Gets metadata and actions for a specific file or directory.