MCP Development Server

You can run a Python-based MCP server that gives Claude complete project context, manages files and templates, integrates with Git, and executes code in Docker environments. This server enables you to automate project setup, tracking, and testing workflows with reliable, isolated execution contexts.

How to use

Connect your MCP client to the local MCP development server to access project structure, status information, and build/test capabilities. Use the server to initialize projects, run build and test operations, manage Docker containers, and query analysis prompts for development suggestions or error diagnosis. All actions are performed through the MCP client’s requests to the server, which returns structured results, artifacts, and status updates.

How to install

Prerequisites you need before installing the MCP server are Python 3.12 or higher, Docker, and Git.

Install the MCP server package from PyPI to get started quickly.

pip install mcp-dev-server

For development setup, clone the repository, install editable mode, and prepare your environment.

git clone https://github.com/your-org/mcp-dev-server.git
cd mcp-dev-server
pip install -e .

Verify installation by running tests or starting a development session as needed.

Configuration

Set up the local Claude Desktop to point at your MCP server. Add the following configuration so your Claude client can launch and communicate with the server.

{
  "mcpServers": {
    "dev": {
      "command": "mcp-dev-server",
      "args": []
    }
  }
}

Notes

The server exposes capabilities for resources (project structure, build status, test results, Docker container status), tools (project initialization, build operations, test execution, Docker commands), and prompts (project analysis, development suggestions, error diagnosis). Use the MCP client to request these capabilities and interpret the results in your workflows.

Development

When developing against this MCP server, ensure you keep your Python environment isolated, use Docker for execution contexts, and run tests with your preferred Python testing tool.

Contributing

If you want to contribute, follow typical open-source contribution practices: create a feature branch, write tests, run the test suite, and submit a pull request with a clear description of changes.

License

The project is licensed under the MIT License. Review the LICENSE file for details.

Available tools

project_init

Initializes a new project within the MCP server, setting up the project structure, templates, and versioned files.

build_operations

Performs build steps, collects artifacts, and reports build status for each project.

test_execution

Runs test suites, aggregates results, and surfaces failures with diagnostic details.

docker_commands

Manages Docker containers, including starting, stopping, and inspecting container status for development tasks.