Claude Dev Server

A Model Context Protocol (MCP) server that enables Claude to assist in software development by providing direct file system access within a specified workspace.

Features

🔍 Read file contents
✍️ Write and modify files
📁 Create directories
📋 List files and directories
ℹ️ Get file information
🛠️ Implement code artifacts directly from Claude

Installation

Make sure you have Python 3.10 or higher installed
Create a new project directory:

mkdir mcp-claude-dev
cd mcp-claude-dev

Set up a virtual environment using UV:

uv venv

Install the package in editable mode:

uv pip install -e .

Project Structure

mcp-claude-dev/
├── claude_dev_server/
│   ├── __init__.py
│   ├── __main__.py
│   ├── server.py
│   └── file_manager.py
└── pyproject.toml

Configuration

To use the server with Claude Desktop, add it to your Claude Desktop configuration file:

Windows: %APPDATA%\Claude\claude_desktop_config.json
macOS: ~/Library/Application Support/Claude/claude_desktop_config.json

Add this configuration:

{
  "mcpServers": {
    "claude-dev": {
      "command": "uv",
      "args": [
        "--directory",
        "PATH_TO_YOUR_PROJECT",
        "run",
        "-m",
        "claude_dev_server",
        "PATH_TO_YOUR_WORKSPACE"
      ]
    }
  }
}

Replace:

PATH_TO_YOUR_PROJECT with the absolute path to your mcp-claude-dev directory
PATH_TO_YOUR_WORKSPACE with the absolute path to the directory where Claude should have file access

Usage

Start Claude Desktop
The server will appear in the tools menu (hammer icon)
You can now ask Claude to:
- Read files: "Can you read the contents of file.txt?"
- Write files: "Create a new file called example.py with a basic Flask app"
- List directories: "What files are in the current directory?"
- Create directories: "Create a new directory called 'src'"
- Get file info: "What's the size and modification date of main.py?"
- Implement code: "Write a Python function to calculate Fibonacci numbers and save it in math_utils.py"

Security Notes

The server only allows access to files within the specified workspace directory
All file operations require explicit paths relative to the workspace
Directory traversal attempts outside the workspace are blocked
File watching is implemented for change detection

Development

To contribute to the project:

Clone the repository
Install development dependencies
Make your changes
Test thoroughly before submitting a pull request

Running Tests

python -m pytest tests/

Available Tools

The server exposes these MCP tools:

read_file(path: str) -> str: Read file contents
write_file(path: str, content: str) -> str: Write content to file
list_files(path: str = "") -> str: List directory contents
create_directory(path: str) -> str: Create new directory
get_file_info(path: str) -> str: Get file metadata
implement_artifact(path: str, content: str) -> str: Save code artifacts

Error Handling

The server includes robust error handling for:

Invalid paths
Access attempts outside workspace
File operation failures
Permission issues

License

MIT License - feel free to use and modify as needed.

Contributing

Contributions are welcome! Please read our contributing guidelines and submit pull requests for any improvements.